JP6283297B2 - Method, apparatus and system for resizing and restoring original depth frame - Google Patents

Description

The present invention relates to a method, apparatus, and system for resizing and restoring an original depth frame.
In the notation of the pixel arrangement in the present invention, “row” refers to a vertical alignment, and “column” refers to a horizontal alignment.

  With the advancement of science and technology, the performance and quality of electronic products have been increasingly required. Among them, when installing the display device, a clear and easy-to-read display screen is provided, and each manufacturer announces a three-dimensional (three-dimensional) display device one after another to give consumers a different vision. Providing an effect.

  At this stage, the 3D image generation system generates a depth frame corresponding to the color frame based on the original 2D image data, that is, a color frame, and passes through an existing transfer device. Color frames and depth frames are transferred to the user's 3D display device. The 3D display device receives a color frame and a depth frame and performs an operation to obtain two or more multi-view images that can be used by all naked-eye 3D display devices, or to the left of the glasses-type 3D display device. An eye image and a right eye image can be obtained.

  However, in the conventional technology, before transferring the above-described color frame and depth frame, the data amount is not reduced or the processing for enhancing the compression efficiency is not performed, and data when transferring the color frame and the depth frame is not performed. Double the amount or processing time.

  This will be described with reference to FIG. FIG. 1 is a diagram illustrating a pixel data structure when a depth frame is transferred in an original state (unprocessed state). Here, as an example, the original depth frame O is a gray scale frame, and each pixel in the color frame corresponding to the original depth frame O has three sub-pixels (sub-pixels). Therefore, the number of sub-pixels of each pixel in the original depth frame O is also three, and these three sub-pixels are red, green, and blue (R, G, B), respectively. However, in practice, since the original depth frame O is a grayscale frame, each subpixel also has a subpixel value (that is, a grayscale value), and these subpixel values are all within each pixel. The same.

  In the above system, how to provide an effective frame compression and decompression method and apply it to a 3D video compression and decompression system to reduce the burden of existing transfer devices and bandwidth Is an important issue.

  In view of the above-described problems, the present invention is applied to a 3D video compression and decompression system, and has the advantages of high compression efficiency, small amount of data, and high transfer efficiency. It is an object of the present invention to provide an effective original depth frame compression and decompression method, apparatus, and system that can reduce the burden of bandwidth.

In order to achieve the above object, a method of changing the size of an original depth frame according to the present invention to a resized depth frame includes obtaining two sub-pixel values of at least two pixels in the original depth frame, respectively. The step of storing the plurality of subpixel values corresponding to two subpixels of one pixel in the resized depth frame is provided.
In one embodiment, the original depth frame is a gray scale frame, and the sub-pixel value of each pixel in the gray scale frame is the same.
In one embodiment, the number of subpixels of each pixel in the original depth frame is three, and the subpixel values of the subpixels are the same.
In one embodiment, the number of sub-pixels of each pixel in the resized depth frame is three, and the step of storing the plurality of sub-pixel values includes three sub-pixel values, one for each of the re-sized depth frames. It is a step of storing corresponding to the three sub-pixels of the pixel.
In one embodiment, the size along the row direction of the resized depth frame is one third of the original depth frame.

In order to achieve the above object, an apparatus for changing a size of an original depth frame to a resized depth frame according to the present invention includes a pixel acquisition unit and a pixel storage unit. The pixel acquisition unit acquires two sub-pixel values of at least two pixels in the original depth frame. The pixel storage unit stores each of the plurality of subpixel values corresponding to two subpixels of one pixel in the resized depth frame.
In one embodiment, the original depth frame is a gray scale frame, and the sub-pixel value of each pixel in the gray scale frame is the same.
In one embodiment, the number of subpixels of each pixel in the original depth frame is three, and the subpixel values of the subpixels are the same.
In one embodiment, the number of sub-pixels of each pixel in the resized depth frame is three, and the pixel storage unit converts three sub-pixel values into three sub-pixels of one pixel in the re-sized depth frame, respectively. It is characterized by being stored correspondingly.
In one embodiment, the size along the row direction of the resized depth frame is one third of the original depth frame.

In order to achieve the above object, a system for changing a size of an original depth frame to a resized depth frame according to the present invention includes a memory unit and a processing unit. The memory unit stores the original depth frame. The processing unit acquires two subpixel values of at least two pixels in the original depth frame, respectively, and stores these subpixel values corresponding to two subpixels of one pixel in the resized depth frame, respectively. It is characterized by that.
In one embodiment, the original depth frame is a gray scale frame, and the sub-pixel value of each pixel in the gray scale frame is the same.
In one embodiment, the number of subpixels of each pixel in the original depth frame is three, and the subpixel values of the subpixels are the same.
In one embodiment, the number of subpixels of each pixel in the resized depth frame is three, and the processing unit obtains three subpixel values of three pixels in the original depth frame, respectively, One subpixel value is stored corresponding to each of the three subpixels of one pixel in the resized depth frame.
In one embodiment, the size along the row direction of the resized depth frame is one third of the original depth frame.

To achieve the above object, a method for restoring a resized depth frame to an original depth frame according to the present invention includes: obtaining a first subpixel value of one pixel in the resized depth frame; Storing in all sub-pixels of the first pixel in the original depth frame; obtaining a second sub-pixel value of the pixel in the resized depth frame; and second pixels in the original depth frame. And storing in all of the sub-pixels.
In one embodiment, the number of sub-pixels of each pixel in the resized depth frame is three, and the step of storing the first sub-pixel value includes the first sub-pixel value in the original depth frame. Storing the second subpixel value in three subpixels of one pixel, and storing the second subpixel value in the three subpixels of the second pixel in the original depth frame. It is a step to be preserved.
In one embodiment, the number of subpixels of each pixel in the original depth frame is three, and the subpixel values of the subpixels are the same.
In one embodiment, the first pixel and the second pixel in the original depth frame are located in the same row of the original depth frame.
In one embodiment, the first pixel and the second pixel in the original depth frame are located in the same column of the original depth frame.
In one embodiment, the size of the original depth frame along the row direction is three times that of the resized depth frame.

In order to achieve the above object, an apparatus for restoring a resized depth frame to an original depth frame according to the present invention includes a pixel acquisition unit and a pixel storage unit. The pixel acquisition unit acquires at least one first subpixel value and second subpixel value of one pixel in the resized depth frame. The pixel storage unit stores the first subpixel value in all subpixels of the first pixel in the original depth frame, and stores the second subpixel value in all subpixels of the second pixel in the original depth frame. It is characterized by making it.
In one embodiment, the number of sub-pixels of each pixel in the resized depth frame is three, and the pixel storage unit uses the first sub-pixel value as three sub-pixels of the first pixel in the original depth frame. The second subpixel value is stored in three subpixels of the second pixel in the original depth frame while being stored in the pixel.
In one embodiment, the number of subpixels of each pixel in the original depth frame is three, and the subpixel value of each of the pixels is the same.
In one embodiment, the first pixel and the second pixel in the original depth frame are located in the same row of the original depth frame.
In one embodiment, the first pixel and the second pixel in the original depth frame are located in the same column of the original depth frame.
In one embodiment, the size of the original depth frame along the row direction is three times that of the resized depth frame.

To achieve the above object, a system for restoring a resized depth frame to an original depth frame according to the present invention includes a memory unit and a processing unit. The memory unit stores the resized depth frame. The processing unit obtains at least one first subpixel value and second subpixel value of one pixel in the resized depth frame, and uses the first subpixel value in all subpixels of the first pixel in the original depth frame. And the second subpixel value is stored in all the subpixels of the second pixel in the original depth frame.
In one embodiment, the number of sub-pixels of each pixel in the resized depth frame is three, and the processing unit uses the first sub-pixel value as the three sub-pixels of the first pixel in the original depth frame. And storing the second subpixel value in three subpixels of the second pixel in the original depth frame.
In one embodiment, the number of subpixels of each pixel in the original depth frame is three, and the subpixel values of the subpixels are the same.
In one embodiment, the first pixel and the second pixel in the original depth frame are located in the same row of the original depth frame.
In one embodiment, the first pixel and the second pixel in the original depth frame are located in the same column of the original depth frame.
In one embodiment, the size of the original depth frame along the row direction is three times that of the resized depth frame.

  As described above, in the method, apparatus, and system for changing the size of the original depth frame to the resized depth frame according to the present invention, the two sub-pixel values of at least two pixels in the original depth frame are respectively acquired, Are stored corresponding to two subpixels of one pixel in the resized depth frame. Therefore, since the data amount of the resized depth frame can be reduced to only one third of the data amount of the original depth frame, the present invention is an effective depth frame compression technique and can be applied to a 3D video compression system, and the compression efficiency is high. It has the advantages of being high, having a small amount of data, and having high transfer efficiency. When applied to the transfer, it is possible to effectively reduce the bandwidth burden of the existing transfer device.

  In the method, apparatus, and system for restoring a resized depth frame to an original depth frame according to the present invention, a step of acquiring a first subpixel value of one pixel in the resized depth frame, and a first subpixel value as an original depth Storing in all sub-pixels of the first pixel in the frame; obtaining a second sub-pixel value of the pixel in the resized depth frame; and subtracting the second sub-pixel value from all of the second pixels in the original depth frame. Storing in the sub-pixels. Therefore, the method, apparatus, and system for restoring a resized depth frame to an original depth frame can be applied to a 3D video restoration system, unlike the conventional technique.

It is the figure which showed the pixel data structure when a depth frame is transferred by the original format (unprocessed state). FIG. 6 is a flowchart illustrating a method for changing a size of an original depth frame to a resized depth frame according to a preferred embodiment of the present invention. FIG. 5 is a block diagram illustrating functions of an apparatus for changing a size of an original depth frame to a resized depth frame according to a preferred embodiment of the present invention. FIG. 5 is a diagram illustrating an original depth frame in a preferred embodiment of the present invention. It is the figure which showed the resized depth frame after the change in the preferable Example of this invention. It is the figure which showed the resized depth frame after the change in the preferable Example of this invention. FIG. 5 is a diagram illustrating an image of an original depth frame in a preferred embodiment of the present invention. FIG. 4B is a diagram showing an image of a resized depth frame after the one shown in FIG. 4A is changed by the method according to the preferred embodiment of the present invention. It is the figure which showed the image | video of the other resized depth frame after changing what is shown to FIG. 4A by the method which concerns on the suitable Example of this invention. FIG. 6 is a diagram illustrating another original depth frame in a preferred embodiment of the present invention. It is the figure which showed the resized depth frame after the other change in the preferable Example of this invention. It is the figure which showed the resized depth frame after the other change in the preferable Example of this invention. It is the figure which showed the image | video of the other original depth frame in the preferable Example of this invention. It is the figure which showed the image | video of the resized depth frame after changing what is shown to FIG. 6A by the method which concerns on the preferable Example of this invention. It is the figure which showed the image | video of the other resized depth frame after changing what was shown to FIG. 6A by the method which concerns on the suitable Example of this invention. FIG. 3 is a block diagram illustrating a function of a system for changing a size of an original depth frame to a resized depth frame in a preferred embodiment of the present invention. FIG. 5 is a flowchart illustrating a method for restoring a resized depth frame to an original depth frame according to an exemplary embodiment of the present invention. FIG. 3 is a block diagram illustrating functions of an apparatus for restoring a resized depth frame to an original depth frame according to a preferred embodiment of the present invention. FIG. 4 is a diagram illustrating a resize depth frame according to a preferred embodiment of the present invention. FIG. 6 is a diagram illustrating an original depth frame after restoration in a preferred embodiment of the present invention. FIG. 6 is a diagram illustrating an original depth frame after restoration in a preferred embodiment of the present invention. It is the figure which showed the image | video of the resized depth frame in the preferable Example of this invention. It is the figure which showed the image | video of the original depth frame after decompress | restoring by the method which concerns on the suitable Example shown above-mentioned FIG. 10B. It is the figure which showed the image | video of the other original depth frame after decompress | restoring by the method which concerns on the suitable Example shown to above-mentioned FIG. 10C. It is the figure which showed the resize depth frame in the other embodiment of this invention. It is the figure which showed the original depth frame in the other embodiment of this invention. It is the figure which showed the original depth frame in the further another embodiment of this invention. It is the figure which showed the image | video of the other resized depth frame in the preferable Example of this invention. It is the figure which showed the image | video of the original depth frame after decompress | restoring by the method which concerns on the suitable Example shown to above-mentioned FIG. 12B. It is the figure which showed the image | video of the other original depth frame after decompress | restoring by the method which concerns on the suitable Example shown to above-mentioned FIG. 12C. FIG. 3 is a block diagram illustrating a function of a system for restoring a resized depth frame to an original depth frame in a preferred embodiment of the present invention.

  Hereinafter, a method, an apparatus, and a system in a preferred embodiment of the present invention will be described with reference to the drawings. Of these, the same constituent elements will be described with the same reference numerals.

  Hereinafter, a method, apparatus, and system for changing the size of the original depth frame to make a resized depth frame will be described.

  This will be described with reference to FIGS. 2A and 2B. FIG. 2A is a flow diagram illustrating a method for changing the size of an original depth frame to a resized depth frame in a preferred embodiment of the present invention. FIG. 2B is a block diagram illustrating the function of the apparatus 1 for changing the size of the original depth frame to a resized depth frame in a preferred embodiment of the present invention.

  As shown in FIG. 2A, the method of changing the size of the original depth frame according to this embodiment to a resized depth frame includes steps S01 to S02. Of these, step S01 is a step of acquiring two sub-pixel values of at least two pixels in the original depth frame. More specifically, this is a step of obtaining at least two subpixel values in total, at least one from each of at least two pixels in the original depth frame. Step S02 is a step of storing these sub-pixel values corresponding to the two sub-pixels of one pixel in the resize depth frame. More specifically, step S02 is a step of storing the at least two subpixel values acquired in step S01 in correspondence with at least two subpixels in one pixel in the resized depth frame. Here, the original depth frame is a gray scale frame obtained by using a depth frame camera, or an original multiview color frame (original multiview color frame, which is a plurality of 2D color images with different viewing angles. This is an original depth frame generated based on a frame). For this reason, the original depth frame and the original color frame are gray scale frames corresponding to each other. In step S01, two pixels in the original depth frame are located in the same row of the original depth frame, or two pixels in the original depth frame are located in the same column of the original depth frame. This will be described in detail below.

  As shown in FIG. 2B, the device 1 (hereinafter referred to as “device 1”) that changes the size of the original depth frame to a resized depth frame includes a pixel acquisition unit 11, a pixel storage unit 12, and a control. A unit 13 is provided. The control unit 13 is electrically connected to the pixel acquisition unit 11 and the pixel storage unit 12 respectively. Among these, the pixel acquisition unit 11 acquires two sub-pixel values of at least two pixels in the original depth frame. Then, the pixel storage unit 12 stores these subpixel values corresponding to the two subpixels of one pixel in the resize depth frame. Here, the functions of the pixel acquisition unit 11 and the pixel storage unit 12 can be realized by software, or the functions of the pixel acquisition unit 11 and the pixel storage unit 12 can be realized by hardware or firmware. The invention is not limited to these. The control unit 13 is a core control component of the apparatus 1 and is composed of, for example, a CPU (central processing unit) or other control hardware, software, or firmware. Among these, the control unit 13 controls the pixel acquisition unit 11 to acquire the sub-pixel value of the original depth frame, and controls the pixel storage unit 12 to set the sub-pixel value in the corresponding sub-pixel of the resized depth frame. It can be saved. Hereinafter, the above-described method and apparatus 1 and its detailed operation process will be described with reference to FIGS. 3A, 3B and 3C.

  This will be described with reference to FIGS. 3A and 3B. FIGS. 3A and 3B are views showing an original depth frame O and a resized depth frame D after change in a preferred embodiment of the present invention, respectively. Here, the original depth frame O is, for example, the same as the original depth frame O of FIG. 1, that is, can be a gray scale frame. In addition, the number of subpixels for each pixel is three, that is, three subpixels (R, G, and B) of red, green, and blue. Since the original depth frame O is a grayscale frame, the three subpixel values (that is, grayscale frame values) of the subpixels are all the same. Further, the number of sub-pixels for each pixel of the resized depth frame D after the change is also three. The original depth frame O of this embodiment is an example in which the pixels include 3 rows vertically and 6 columns horizontally, that is, 3 × 6 = 18 pixels. In this case, since each pixel has three subpixels, the number of subpixels has a total of 54 subpixels of 9 rows × 6 columns. Hereinafter, the position of each sub-pixel and its sub-pixel value are displayed by code. For example, the subpixels in the first row and the first column shown in FIG. 3A are displayed as R (30). This means that the sub-pixel is the sub-pixel R, 30 is the sub-pixel value, and (1, 1) represents the position. The subpixels in the fifth row and the third column are displayed as G (33). This means that the sub-pixel is the sub-pixel G, 33 is the sub-pixel value, and its position is represented by (5, 3). Other sub-pixels follow this display method.

  In the frame resizing method of the present embodiment, three subpixel values of three pixels in the original depth frame O are acquired through the pixel acquisition unit 11, and these subpixel values are respectively acquired by the pixel storage unit 12. The resize depth frame D is stored corresponding to the three sub-pixels of one pixel. That is, more specifically, one subpixel value is acquired from each of the three pixels in the original depth frame O through the pixel acquisition unit 11, that is, a total of three subpixel values are acquired, The storage unit 12 stores these subpixel values corresponding to the three subpixels in one pixel of the resized depth frame D, respectively. As shown in FIGS. 3A and 3B, in this embodiment, the sub-pixel value 30 at the position (1, 1) (sub-pixel R) in the pixel of the original depth frame O in FIG. Since all three subpixel values in each pixel in frame O are the same, the subpixel value at position (2, 1) (subpixel G) or position (3, 1) (subpixel B) is acquired. The subpixel value 30 is stored at the position (1, 1) (subpixel R) of the resized depth frame D in FIG. 3B. The sub-pixel value 31 at the position (1, 2) (sub-pixel R) in the other pixels in the original depth frame O is acquired, and this sub-pixel value 31 is obtained as the position (2, 1) (sub-pixel) of the resized depth frame D. Save in pixel G). The sub-pixel value 32 at the position (1, 3) (sub-pixel R) in still another pixel of the original depth frame O is obtained, and this sub-pixel value 32 is obtained at the position (3, 1) ( Stored in sub-pixel B). The sub-pixel value 34 at the position (1, 4) (sub-pixel R) of the further pixel of the original depth frame O is acquired, and this sub-pixel value 34 is obtained at the position (1, 2) (sub-pixel) of the resized depth frame D. R). A sub-pixel value 36 of still another pixel position (1, 5) (sub-pixel R) of the original depth frame O is acquired, and this sub-pixel value 36 is obtained as (2, 2) (sub-pixel) of the resized depth frame D. G). The sub-pixel value 35 of still another pixel position (1, 6) (sub-pixel R) of the original depth frame O is acquired, and the sub-pixel value 35 is obtained at the position (3, 2) (sub-position of the resized depth frame D. Save in pixel B). The sub-pixel value 29 of still another pixel position (4, 1) (sub-pixel R) of the original depth frame O is acquired, and this sub-pixel value 29 is obtained as the position (4, 1) (sub-pixel) of the resized depth frame D. Save in pixel R). The sub-pixel value 30 of the further pixel position (4, 2) (sub-pixel R) of the original depth frame O is acquired, and this sub-pixel value 30 is converted into the position (5, 1) (sub-pixel G of the resized depth frame D. ). (Repeat the same operation below.) The sub-pixel value 103 of still another pixel position (7, 4) (sub-pixel R) of the original depth frame O is acquired, and the sub-pixel value 103 is obtained as the position (7, 2) (sub-pixel) of the resized depth frame D. Save in pixel R). The sub-pixel value 104 of still another pixel position (7, 5) (sub-pixel R) in the original depth frame O is acquired, and this sub-pixel value 104 is obtained at the position (8, 2) (sub-pixel) of the resized depth frame D. Save in pixel G). The sub-pixel value 106 of still another pixel position (7, 6) (sub-pixel R) of the original depth frame O is acquired, and the sub-pixel value 106 is obtained as the position (9, 2) (sub-pixel) of the resized depth frame D. Save in pixel B). Therefore, the size along the row direction (vertical direction) D2 of the resized depth frame D of the present embodiment is at least one third of the original depth frame O (here, a case of one third is taken as an example). However, the size along the column direction (lateral direction) D1 is the same as that of the original depth frame O. Therefore, the data amount of the resized depth frame D is an amount that is at least one third of the original depth frame O (here, the case of one third is taken as an example). It should be noted that the order of obtaining the above subpixel values is merely an example, and in other embodiments, in order to increase the accuracy after resizing, the above subpixel values may be obtained and stored. The order may be different from the above (or the relative positional relationship after resizing may be considered different). For example, as shown in FIG. 3C, the first column is the same as the arrangement of the first column in FIG. 3B. However, the pixels in the second column are arranged by moving the sub-pixel values for each pixel in the second column in FIG. 3B. Among these, in each pixel of FIG. 3B, for example, the subpixel value 34 at the position (1, 2) is moved to the right by one, and the subpixel value 36 at the position (2, 2) is moved to the right by one. Then, the sub-pixel value 35 at the position (3, 2) has been moved to the left by the first column and the second row. The arrangement of sub-pixel values in other pixels is also the same. The rearrangement and movement of the subpixel values described above are merely examples, and the present invention is not limited to this. Thereby, the accuracy of the original depth frame can be maintained even after resizing by changing the arrangement position of the sub-pixel values of the resized depth frame corresponding to the original depth frame or rearranging the order.

  This will be described with reference to FIGS. 4A, 4B, and 4C. FIG. 4A is a diagram illustrating an image of an original depth frame in a preferred embodiment of the present invention. 4B and 4C are views showing images of the resized depth frame after the one shown in FIG. 4A is changed by the method of the embodiment of the present invention. 4B corresponds to FIG. 3B, and FIG. 4C corresponds to FIG. 3C. As such, FIGS. 4B and 4C correspond to the methods of FIGS. 3B and 3C, respectively, so the images shown are different, but as far as the naked eye is concerned, the images of FIGS. 4B and 4C are very similar. It is not easy to distinguish.

  The size (analysis level) of the gray scale frame in FIG. 4A is 1920 × 810, and the size shown in FIGS. 4B and 4C is 1920 × 270 (270 = 810 × 1/3). For this reason, the size of the resized depth frame of FIG. 4B and FIG. 4C is 1/3 of the original depth frame O of FIG. 4A. Therefore, the present invention is an effective depth frame compression technique, and can be applied to a 3D video compression system, and has advantages such as high compression efficiency, a small amount of data, and high transfer efficiency. In addition, when applied to transfer, it is possible to reduce the bandwidth burden of an existing transfer device. At the same time, in this embodiment, the sub-pixel value can be processed through a change in position or a change in order, so that the method of changing the size of the original depth frame to a resized depth frame is another compression method. Suitable for use in combination with technology, depth frame accuracy is maintained even if frame compression is performed twice.

  Next, a description will be given with reference to FIGS. 5A and 5B. FIGS. 5A and 5B are views showing an original depth frame O and a resized depth frame D according to another embodiment of the present invention, respectively. Again, the original depth frame O is a gray scale frame, and the number of subpixels for each pixel is three (R, G, B) as an example. Furthermore, the three subpixel values (grayscale values) of each subpixel are all the same. Further, the number of sub-pixels for each pixel of the resized depth frame D is also three. The original depth frame O of the present embodiment is also an example in which pixels include 3 rows vertically and 6 columns horizontally, that is, 3 × 6 = 18 pixels. Therefore, the sub-pixel has a total of 54 sub-pixels of 9 rows × 6 columns. In the following, the position of each subpixel and its subpixel value (grayscale value) are displayed by the code and the number. For example, the subpixel in the first row and the first column in FIG. 5A is displayed as R (195), meaning that this subpixel is the subpixel R, and 195 is its subpixel value, (1, The position is represented by 1). The subpixel in the third row and the fifth column is displayed as B (26), which means that this subpixel is the subpixel B. 26 is the subpixel value, and (3,5) indicates its position. Represents. Other sub-pixels follow this display method.

  In this embodiment, as shown in FIG. 5A and FIG. 5B, the sub-pixel value 195 at the position (1, 1) in the pixel of the original depth frame O is acquired and [three pixels of each pixel in the original depth frame O are obtained. Since all the subpixel values are the same, it is also possible to acquire the subpixel value in the position (2, 1) or (3, 1)]. This subpixel value 195 is the position of the resized depth frame D. Save in (1,1). The sub-pixel value 198 at the position (4, 1) in the other pixels of the original depth frame O is acquired, and the sub-pixel value 198 is stored in the position (2, 1) of the resized depth frame D. The sub-pixel value 200 at the position (7, 1) in still another pixel of the original depth frame O is acquired, and the sub-pixel value 200 is stored in the position (3, 1) of the resized depth frame D. The sub-pixel value 190 at the position (1, 2) in still another pixel of the original depth frame O is acquired, and the sub-pixel value 190 is stored in the position (1, 2) of the resized depth frame D. (Repeat the same operation below.) The sub-pixel value 31 of still another pixel position (4, 6) of the original depth frame O is acquired, and the sub-pixel value 31 is stored in the position (2, 6) of the resized depth frame D. The sub-pixel value 35 of the further pixel position (7, 6) of the original depth frame O is acquired, and the sub-pixel value 35 is stored in the position (3, 6) of the resized depth frame D. Therefore, the size along the horizontal column direction D1 of the resized depth frame D of the present embodiment is one third of the original depth frame O, but the size along the row direction D2 is the same as the original depth frame O. Therefore, the data amount of the resized depth frame D is only one third of that of the original depth frame O. It should be noted that the above-described subpixel value acquisition order is also merely an example, and in other embodiments, the subpixel value acquisition order may be different in order to improve accuracy. For example, as shown in FIG. 5C, the arrangement of the first horizontal row and the first horizontal row of FIG. 5B are the same. In the second column, the sub-pixel values in the second column in FIG. 5B are shifted by one, and in the third column, the sub-pixel values in the third column in FIG. 5B are moved by two. It is an arrangement. In addition, the fourth column in FIG. 5C has the same arrangement as the fourth column in FIG. 5B. In the fifth column, the sub-pixel values in the fifth column in FIG. 5B are shifted by one and arranged. In the sixth column, the sub-pixel values in the sixth column in FIG. 5B are shifted and arranged. The above-described rearrangement and movement of the sub-pixel values are merely examples, and do not limit the present invention.

  Next, a description will be given with reference to FIGS. 6A, 6B, and 6C. FIG. 6A is a diagram illustrating an image of an original depth frame in a preferred embodiment of the present invention. FIG. 6B and FIG. 6C are views showing images of the resized depth frame after the one shown in FIG. 6A is changed by the method of the embodiment of the present invention. Of these, FIGS. 6B and 6C correspond to the methods shown in FIGS. 5B and 5C, respectively. As such, FIGS. 6B and 6C correspond to the methods of FIGS. 5B and 5C, respectively, so the images shown are different, but as far as the naked eye is concerned, FIG. 6B is very close to the images of FIG. 6C. It is not easy to distinguish. The size (analysis level) of the gray scale frame in FIG. 6A is 1920 × 810, and the size in FIG. 6B and FIG. 6C is 630 (630 = 1920 × 1/3) × 810, so the resizing in FIG. 6B and FIG. The depth frame size is one third of the original depth frame O in FIG. 6A.

  It should be noted again that the size of the original depth frame O described above and each pixel has three subpixels, but this is merely an example, and of course, in other embodiments, the original depth frame O It is also possible that the size and the number of sub-pixels are different. For example, the number of subpixels may be two, four or more (including four), and the present invention is not limited to this.

  The description will be made with reference to FIG. FIG. 7 is a block diagram showing the function of the system 2 (hereinafter abbreviated as “system 2”) that changes the size of the original depth frame to a resized depth frame in a preferred embodiment of the present invention.

  The system 2 includes a memory unit 21 and a processing unit 22. The memory unit 21 is electrically connected to the processing unit 22. The memory unit 21 stores the original depth frame. Of course, the memory unit 21 can also store the resized depth frame. The memory unit 21 is a non-transitory computer readable storage medium. For example, a memory, a memory card, a CD-ROM, a video tape, a magnetic tape for computers, or any combination thereof is used for storing information. Among these, the memory is ROM, RAM, flash memory, FPGA (Field-Programmable Gate Array), or other types of memory, but is not particularly limited here.

  The processing unit 22 can comprise the core control components of the system 2. For example, it includes at least one CPU and memory, or other hardware, software, or firmware for controlling. The processing unit 22 acquires two sub-pixel values of at least two pixels in the original depth frame stored by the memory unit 21, and uses the acquired sub-pixel values for two pixels of one pixel in the resized depth frame. It can be stored correspondingly in the sub-pixel. More specifically, the processing unit 22 acquires a total of at least two subpixel values, at least one from each of at least two pixels in the original depth frame stored by the memory unit 21, and acquires these subpixels. Each value can be stored correspondingly in at least two sub-pixels in one pixel in the resized depth frame. In that case, as shown in FIGS. 3B and 3C, the size along the row direction D2 of the resized depth frame is set to one third of the original depth frame, or as shown in FIGS. 5B and 5C, The size along the column direction D1 of the depth frame can be set to one third of the original depth frame.

  Further, in the system 2, the other technical features related to changing the size of the original depth frame to the resized depth frame conform to the above-described resizing method and the technical content of the apparatus 1, and therefore are described here again. do not do.

  Next, a method, apparatus and system for restoring a resized depth frame to an original depth frame will be described.

  This will be described with reference to FIGS. FIG. 8 is a flowchart illustrating a method for restoring a resized depth frame to an original depth frame in a preferred embodiment of the present invention. FIG. 9 is a block diagram showing the function of the device 3 for restoring the resized depth frame to the original depth frame in the preferred embodiment of the present invention.

  As shown in FIG. 8, the method for restoring the resized depth frame to the original depth frame according to the present embodiment includes steps P01 to P04. Of these, step P01 is a step of acquiring the first sub-pixel value of one pixel in the resized depth frame. Step P02 is a step of storing the first subpixel value in all the subpixels of the first pixel in the original depth frame. Step P03 is a step of acquiring the second sub-pixel value of the pixel (the “one pixel”) in the resized depth frame. Step P04 is a step of storing the second subpixel value in all the subpixels of the second pixel in the original depth frame. Here, the resized depth frame can be acquired based on the original multi-view color frame. In Step P01 and Step P02, the first pixel and the second pixel of the original depth frame are located in the same row of the original depth frame or in the same column of the original depth frame. This will be described in detail below.

  Also, as shown in FIG. 9, the device 3 that restores the resized depth frame to the original depth frame (hereinafter referred to as “device 3”) includes a pixel acquisition unit 31, a pixel storage unit 32, and a control unit 33. The control unit 33 is electrically connected to the pixel acquisition unit 31 and the pixel storage unit 32, respectively. Among these, the pixel acquisition unit 31 acquires at least one first subpixel value and second subpixel value of one pixel in the resize depth frame. The pixel storage unit 32 stores the first subpixel value acquired by the pixel acquisition unit 31 in all the subpixels of the first pixel in the original depth frame, and also stores the second subpixel value in the second depth in the original depth frame. Save in all sub-pixels of the pixel. Here, the functions of the pixel acquisition unit 31 and the pixel storage unit 32 are realized by software, or the functions of the pixel acquisition unit 31 and the pixel storage unit 32 are realized by hardware or firmware. do not do. The control unit 33 includes the core control component of the device 3. For example, it is composed of a CPU or other control hardware, software or firmware. Among these, the control unit 33 controls the pixel acquisition unit 31 to acquire the sub-pixel value of each pixel in the resized depth frame, and the acquired sub-pixel value is stored in all the corresponding sub-pixels of the original depth frame. The pixel storage unit 32 is controlled so as to be stored. Hereinafter, the above-described restoration method and apparatus 3 and the detailed operation process thereof will be described with reference to FIGS. 10A, 10B, and 10C.

  This will be described with reference to FIGS. 10A and 10B. FIGS. 10A and 10B are views showing a resized depth frame D and an original depth frame O, respectively, according to a preferred embodiment of the present invention. Here, the resized depth frame D is a gray scale frame, and the case where the number of sub-pixels for each pixel is three is an example, and each pixel has three sub-pixels (R, G, B). The number of sub-pixels for each pixel of the original depth frame O is also three (each pixel also has three sub-pixels R, G, and B, respectively). The resized depth frame D of FIG. 10A of the present embodiment is an example in which 3 rows in the vertical direction and 2 columns in the horizontal direction, that is, 3 × 2 = 6 pixels are included. As described above, each pixel has 3 sub-pixels, so that it has a total of 18 sub-pixels of 9 rows × 2 columns. In the same manner as described above, the position of each subpixel and its subpixel value are displayed by code. For example, the subpixel in the first row and the first column in FIG. 10A is represented by R (30), which means that this subpixel is the subpixel R, and 30 represents the subpixel value, (1 , 1) represents the position. The subpixel in the fifth row and the second column is displayed as G (39), which means that this subpixel is the subpixel G, 39 represents its subpixel value, and (5, 2) indicates its position. Represents. Other sub-pixels follow this display method.

  In the frame size restoration method according to the present embodiment, the first subpixel value, the second subpixel value, and the third subpixel value of one pixel in the resize depth frame D are acquired via the pixel acquisition unit 31. The acquired first subpixel value is stored in all the subpixels of the first pixel in the original depth frame via the pixel storage unit 32, and the acquired second subpixel value is stored in the original depth frame. The third subpixel value is stored in all the subpixels of the second pixel, and the acquired third subpixel value is stored in all the subpixels of the third pixel in the original depth frame.

  As shown in FIGS. 10A and 10B, in this embodiment, the first subpixel value 30 at the position (1, 1) (subpixel R) in the pixel in the resized depth frame D via the pixel acquisition unit 31. , The second subpixel value 31 at the position (2, 1) (subpixel G) and the third subpixel value 32 at the position (3, 1) (subpixel B) are obtained, and the pixel storage unit 32 is The first sub-pixel value 30 is changed to position (1, 1) (sub-pixel R), position (2, 1) (sub-pixel G), and position (3, 3) in the first pixel of the original depth frame O, respectively. 1) stored in (sub-pixel B), and the second sub-pixel value 31 is stored in position (1, 2) (sub-pixel R), position (2, 2) in the second pixel of the original depth frame O, respectively. (Sub-pixel G) and position (3, 2) (sub-pixel B) are stored, and the third sub-pixel value 3 In position (1, 3) (subpixel R), position (2, 3) (subpixel G), and position (3, 3) (subpixel B) in the third pixel of the original depth frame O, respectively. Save. Here, the first pixel, the second pixel, and the third pixel (the same subpixel) of the original depth frame O are located in the same row in the vertical direction of the original depth frame O.

  Subsequently, the first subpixel value 34 at the position (1, 2) (subpixel R) and the second subpixel value 36 at the position (2, 2) (subpixel G) in the other pixels of the resized depth frame D, And the third subpixel value 35 of the position (3, 2) (subpixel B) is obtained, and the first subpixel value 34 is respectively set to the position (1, 4) (subpixel) in the fourth pixel of the original depth frame O. Pixel R), position (2,4) (subpixel G), and position (3,4) (subpixel B), and the second subpixel value 36 is stored in the fifth depth of the original depth frame O. The pixel is stored in position (1, 5) (subpixel R), position (2, 5) (subpixel G), and position (3, 5) (subpixel B) in the pixel. The values 35 are the position (1, 6) (subpixel R) and the position (2, 6) (subpixel G) in the sixth pixel of the original depth frame O, respectively. And location (3,6) is stored in a (subpixel B) within.

  Subsequently, the first subpixel value 29 at the position (4, 1) (subpixel R) and the second subpixel at the position (5, 1) (subpixel G) in still another pixel of the resized depth frame D. The value 30 and the third subpixel value 33 at the position (6, 1) (subpixel B) are acquired, and the first subpixel value 29 is set to the position (4, 1) in the seventh pixel of the original depth frame O, respectively. ) (Subpixel R), position (5,1) (subpixel G), and position (6,1) (subpixel B), and the second subpixel value 30 is stored in the original depth frame O. Are stored in position (4,2) (subpixel R), position (5,2) (subpixel G), and position (6,2) (subpixel B) in the eighth pixel, 3 sub-pixel values 33 are set to position (4, 3) (sub-pixel R), position (5, 5) in the ninth pixel of the original depth frame O, respectively. ) (Subpixel G) and the position (6,3) (to be stored in the subpixel B) within. (Repeat the same operation below.) Finally, the first subpixel value 103 at position (7,2) (subpixel R) and the second subpixel value at position (8,2) (subpixel G) in still another pixel of the resized depth frame D. 104 and the third subpixel value 106 at the position (9, 2) (subpixel B) are acquired, and the first subpixel value 103 is respectively set to the position (7, 4) in the sixteenth pixel of the original depth frame O. (Subpixel R), position (8,4) (subpixel G), and position (9,4) (subpixel B) are stored, and the second subpixel value 104 is stored in the original depth frame O, respectively. It is stored in position (7,5) (subpixel R), position (8,5) (subpixel G), and position (9,5) (subpixel B) in the seventeenth pixel, and third The subpixel value 106 is set to the position (7, 6) in the 18th pixel of the original depth frame O (subpixel R The position (8,6) (subpixel G), and the position (9,6) is stored in a (subpixel B) within. Therefore, the size along the vertical row direction D2 of the original depth frame O of the present embodiment is three times the resize depth frame D, but the size along the horizontal column direction D1 is the same as the resize depth frame D. Since they are the same, the data amount of the original depth frame O is three times that of the resized depth frame D in terms of the data amount.

  The subpixel value acquisition order described above is merely an example, and in other embodiments, the subpixel value acquisition order and storage order are different in order to improve accuracy after size restoration (or It is also possible that the relative positional relationship after the change is different). For example, as shown in FIG. 10C, the first column, the second column, and the third column are the same as the arrangement of the first column, the second column, and the third column in FIG. 10B, respectively. The pixels in the fourth, fifth, and sixth columns are arranged by moving one sub-pixel value of each pixel in the fourth, fifth, and sixth columns in FIG. 10B, respectively. Of these, the fourth column in FIG. 10B is the same as the fifth column in FIG. 10C, the fifth column in FIG. 10B is the same as the sixth column in FIG. 10C, and the sixth column in FIG. Same as 4 columns. The above-described rearrangement and movement of the sub-pixel values are merely examples, and do not limit the present invention. Thereby, it is possible to maintain accuracy even after resizing the resized depth frame by changing the arrangement position of the sub-pixel values of the original depth frame corresponding to the resized depth frame or rearranging the order. .

  Next, a description will be given with reference to FIGS. 11A, 11B, and 11C. FIG. 11A is a diagram showing an image of a resized depth frame in a preferred embodiment of the present invention. FIG. 11B is a diagram showing an image of the original depth frame after the one shown in FIG. 11A is restored by the method of the preferred embodiment shown in FIG. 10B. FIG. 11C is a diagram showing an image of the original depth frame after another resized depth frame is restored by the method of the preferred embodiment shown in FIG. 10C. Here, the video of the other resized depth frame before the change in FIG. 11C is not shown because it is similar to that of FIG. 11A. 11B and 11C correspond to the methods shown in FIGS. 10B and 10C, respectively. As such, FIGS. 11B and 11C correspond to the methods of FIGS. 10B and 10C, respectively, so the images shown are different, but as far as the naked eye is concerned, the images of FIGS. 11B and 11C are very similar. It is not easy to distinguish. The size of the gray scale frame of FIG. 11A is 1920 × 270, and the size of FIG. 11B and FIG. 11C is 1920 × 810 (810 = 270 × 3), so the size of the original depth frame of FIG. 11B and FIG. Is three times the resized depth frame of FIG. 11A.

  Further description will be made with reference to FIGS. 12A and 12B. 12A and 12B are views showing a resized depth frame D and a restored original depth frame O according to another embodiment of the present invention, respectively. Here again, the resized depth frame D is a gray scale frame, and the number of subpixels of each pixel is three as an example (each pixel has three subpixels R, G, and B). Further, the number of sub-pixels of each pixel in the original depth frame O is also three (each pixel has three sub-pixels R, G, and B). The resized depth frame D of the present embodiment has a total of 18 sub-pixels of 3 rows × 6 columns, as an example in which 1 × 6 = 6 pixels are provided. Also in the following, the position of each subpixel and its subpixel value are displayed by the code as described above. For example, the subpixel in the first row and the first column in FIG. 12A is displayed as R (195), which means that the subpixel is the subpixel R, and 195 represents the subpixel value. , 1) represents the position. The subpixel in the third row and the second column is displayed as B (202), which means that this subpixel is the subpixel B, 202 represents the subpixel value, and (3, 2) represents the subpixel. Represents the position. Other sub-pixels follow this display method.

  In this embodiment, as shown in FIGS. 12A and 12B, the first subpixel value 195, the position (2,1) (subpixel) of the position (1,1) (subpixel R) in the pixel of the resized depth frame D The second subpixel value 198 of the pixel G) and the third subpixel value 200 of the position (3, 1) (subpixel B) are obtained, and the first subpixel value 195 is obtained as the first pixel of the original depth frame O, respectively. It is stored in position (1,1) (subpixel R), position (2,1) (subpixel G), and position (3,1) (subpixel B), and the second subpixel value 198 in position (4,1) (subpixel R), position (5,1) (subpixel G), and position (6,1) (subpixel B) in the second pixel of the original depth frame O, respectively. In addition, the third subpixel value 200 is stored in the third pixel of the original depth frame O. (7,1) (subpixel R), position (8,1) (subpixel G), and the position (9,1) is stored in a (subpixel B) within. Here, the first pixel, the second pixel, and the third pixel of the original depth frame O are located in the same column of the original depth frame O. Next, in the other pixels of the resized depth frame D, the first subpixel value 190 at the position (1, 2) (subpixel R), the second subpixel value 197 at the position (2, 2) (subpixel G), And the third subpixel value 202 of the position (3, 2) (subpixel B) is obtained, and the first subpixel value 190 is respectively set to the position (1, 2) (subpixel) in the fourth pixel of the original depth frame O. Pixel R), position (2, 2) (subpixel G), and position (3, 2) (subpixel B), and the second subpixel value 197 is stored in the fifth depth of the original depth frame O, respectively. The pixel is stored in position (4, 2) (subpixel R), position (5, 2) (subpixel G), and position (6, 2) (subpixel B) in the pixel, and the third subpixel. The values 202 are respectively set to the position (7, 2) (subpixel R) and the position (8, 2) in the sixth pixel of the original depth frame O ( Pixels G), and the position (9,2) (to be stored in the subpixel B) within. (Repeat the same operation below.) Finally, the first subpixel value 28 at the position (1, 6) (subpixel R) and the second subpixel value 31 at the position (2, 6) (subpixel G) in still another pixel of the resized depth frame D. , And the third subpixel value 35 at the position (3, 6) (subpixel B), and the first subpixel value 28 is set to the position (1, 6) (in the sixteenth pixel of the original depth frame O, respectively. Sub-pixel R), position (2, 6) (sub-pixel G), and position (3, 6) (sub-pixel B), and the second sub-pixel value 31 is stored in the original depth frame O. Stored in position (4, 6) (subpixel R), position (5, 6) (subpixel G), and position (6, 6) (subpixel B) in 17 pixels, and the third sub Pixel value 35 is set to position (7, 6) (subpixel R) and position (8, 6) in the 18th pixel of original depth frame O, respectively. (Subpixel G), and the position (9,6) is stored in a (subpixel B) within. Therefore, the size along the horizontal column direction D1 of the original depth frame O of the present embodiment is three times as large as the resize depth frame D, but the size along the vertical row direction D2 is the same as the resize depth frame D. Since they are the same, the data amount of the original depth frame O is also three times that of the resized depth frame D in terms of the data amount.

  It should be noted that the subpixel value acquisition order described above is merely an example, and in other embodiments, the above subpixel value acquisition and storage is performed in order to improve accuracy after size restoration. Can be different (or the relative positional relationship after the change can be considered different). For example, as shown in FIG. 12C, the first column and the fourth column are the same as the arrangement of the first column and the fourth column in FIG. 12B, respectively, and the second column and the fifth column are respectively shown in FIG. The pixel values in the second column and the fifth column of 12B are arranged by shifting one pixel (three sub-pixels), and the third column and the sixth column are respectively shown in FIG. Each pixel value in the third column and the sixth column is arranged by moving two pixels. As a result, it is possible to maintain accuracy even after resizing the resized depth frame by changing the arrangement position of the sub-pixel values of the original depth frame corresponding to the resized depth frame or rearranging the order. .

  The above-described rearrangement and movement of the sub-pixel values are merely examples, and do not limit the present invention. Also, the size of the above-mentioned resized depth frame D and the fact that each pixel has three pixels are merely examples. Of course, in other embodiments, the size of the resized depth frame D and the number of sub-pixels thereof may be different. For example, the number of subpixels may be two, four or more (including four), and the present invention is not limited to this.

  Next, a description will be given with reference to FIGS. 13A, 13B, and 13C. FIG. 13A is a diagram showing an image of another resized depth frame according to a preferred embodiment of the present invention. FIG. 13B is a diagram showing an image of the original depth frame after the one shown in FIG. 13A is restored by the method of the preferred embodiment shown in FIG. 12B. FIG. 13C is a diagram showing an image of the original depth frame after another resized depth frame is restored by the method of the preferred embodiment shown in FIG. 12C described above. Here, the video of the other resized depth frame before the change in FIG. 13C is similar to that of FIG. 13A and is not shown. Of these, FIGS. 13B and 13C correspond to the methods shown in FIGS. 12B and 12C, respectively. As such, FIGS. 13B and 13C correspond to the methods of FIGS. 12B and 12C, respectively, so the images shown are different, but as far as the naked eye is concerned, the images of FIGS. 13B and 13C are very approximate. It is not easy to distinguish. The size of the gray scale frame of FIG. 13A is 630 × 810, and the size of FIG. 13B and FIG. 13C is 1920 × 810 (1920 = 630 × 3), so the size of the original depth frame of FIG. 13B and FIG. 13C Is three times the resized depth frame of FIG. 13A.

  Next, a description will be given with reference to FIG. FIG. 14 is a block diagram showing the function of the system 4 (hereinafter referred to as “system 4”) for restoring the resized depth frame to the original depth frame in the preferred embodiment of the present invention.

  The system 4 includes a memory unit 41 and a processing unit 42. The memory unit 41 is electrically connected to the processing unit 42. The memory unit 41 stores the resized depth frame. Of course, the memory unit 41 can also store the original depth frame. The memory unit 41 is a non-transitory computer-readable medium, such as a memory, a memory card, a CD-ROM, a video tape, a computer magnetic tape, or any combination thereof, and is used for storing information. It is done. Among these, the memory is ROM, RAM, flash memory, FPGA (Field-Programmable Gate Array), or other types of memory, but is not particularly limited here.

  The processing unit 42 includes the core control components of the system 4. For example, it may comprise at least one CPU and memory, or other control hardware, software or firmware. Among these, the processing unit 42 acquires at least one first subpixel value and second subpixel value of one pixel in the resized depth frame, and uses the first subpixel value for all the first pixels in the original depth frame. The second subpixel value is stored in all the subpixels of the second pixel in the original depth frame. In another embodiment, the first pixel and the second pixel of the original depth frame are located in the same row or column of the original depth frame. The size along the row direction of the original depth frame is three times that of the resized depth frame, or the size along the column direction of the original depth frame is three times that of the resized depth frame.

  Further, in the system 4, other technical features related to restoring the resized depth frame to the original depth frame conform to the above-described restoration method and the technical contents of the apparatus 3, and therefore will not be described again here.

  As described above, in the method, apparatus, and system for changing the size of the original depth frame to the resized depth frame according to the present invention, the two sub-pixel values of at least two pixels in the original depth frame are respectively acquired, Are stored corresponding to the two subpixels of one pixel of the resized depth frame. As a result, the data amount of the resized depth frame is half or smaller than the original depth frame, for example, one third. Therefore, the present invention is an effective depth frame compression technique and can be applied to a 3D video compression system. Further, it has advantages such as high compression efficiency, small amount of data, and high transfer efficiency. When applied to transfer, the effect of reducing the bandwidth burden of the existing transfer device can be achieved.

  In addition, a method, apparatus, and system for restoring a resized depth frame to an original depth frame according to the present invention includes a step of obtaining a first subpixel value of one pixel in the resized depth frame, and the first subpixel value is converted into an original depth frame. Storing in all the sub-pixels of the first pixel in the step, obtaining the second sub-pixel value of the pixel of the resized depth frame, and the second sub-pixel value of all of the second pixels in the original depth frame Storing in the sub-pixel. Accordingly, the method, apparatus, and system for restoring the resized depth frame to the original depth frame can be applied to a 3D video restoration system, unlike the conventional technique.

  As described above, the embodiments of the present invention have been described in detail with reference to the drawings. However, the specific configuration is not limited to these embodiments, and there are design changes and the like without departing from the gist of the present invention. However, it is included in the technical scope of the present invention.

  The present invention provides a method for changing the size of an original depth frame to a resized depth frame by using the above-described configuration, a restoration method thereof, an apparatus and a system corresponding to these methods, and compression and restoration of 3D video. Besides being applicable to the system, it has the advantages of high compression efficiency, small amount of data, and high transfer efficiency, and can effectively reduce the burden of existing transfer devices and bandwidth.

1, 3 Device 11, 31 Pixel acquisition unit 12, 32 Pixel storage unit 13, 33 Control unit 2, 4 System 21, 41 Memory unit 22, 42 Processing unit B, G, R Sub-pixel D Resize depth frame D1 Column direction D2 Row direction O Original depth frame P01-P04, S01-S02 Step

Claims (27)

A method for changing the size of the original depth frame to a resized depth frame,
Obtaining one subpixel value from each of the three pixels in the original depth frame;
Storing the three sub-pixel values of the three pixels in correspondence with the three sub-pixels of the same pixel in the resized depth frame, respectively. How to change to a resized depth frame.   The method of claim 1, wherein the original depth frame is a gray scale frame, and the sub-pixel value of each pixel in the gray scale frame is the same.   The method of claim 1, wherein the number of subpixels of each pixel in the original depth frame is three, and the subpixel values of the subpixels are the same.   The method of claim 1, wherein a size of the resized depth frame along a row direction is one third of the original depth frame. A device that changes the size of the original depth frame to a resized depth frame,
A pixel acquisition unit that acquires one subpixel value from each of the three pixels in the original depth frame;
A pixel storage unit for storing the three sub-pixel values of the three pixels corresponding to the three sub-pixels of the same pixel in the resized depth frame, respectively. A device that changes the size to a resized depth frame.   The apparatus of claim 5, wherein the original depth frame is a gray scale frame, and the sub-pixel value of each pixel in the gray scale frame is the same.   The apparatus according to claim 5, wherein the number of sub-pixels of each pixel in the original depth frame is three, and the sub-pixel values of the sub-pixels are the same.   The apparatus of claim 5, wherein a size of the resized depth frame along a row direction is one third of the original depth frame. A system that changes the size of the original depth frame to a resized depth frame,
A memory unit for storing the original depth frame;
One subpixel value is acquired from each of the three pixels in the original depth frame, and the three subpixel values of the three pixels are respectively associated with the three subpixels of the same pixel in the resized depth frame. And a processing unit for saving the original depth frame to change the size of the original depth frame into a resized depth frame.   The system of claim 9, wherein the original depth frame is a gray scale frame, and the sub-pixel value of each pixel in the gray scale frame is the same.   The system according to claim 9, wherein the number of subpixels of each pixel in the original depth frame is three, and the subpixel values of the subpixels are the same.   The system according to claim 9, wherein a size of the resized depth frame in a row direction is one third of the original depth frame. A method for restoring a resized depth frame to an original depth frame,
Obtaining a first sub-pixel value of one pixel in the resized depth frame;
Storing the first subpixel value in three subpixels of the first pixel in the original depth frame;
Obtaining a second sub-pixel value of the pixel in the resized depth frame;
Storing the second subpixel value in three subpixels of a second pixel in the original depth frame;
Obtaining a third sub-pixel value of the pixel in the resize depth frame;
Storing the third subpixel value in three subpixels of a third pixel in the original depth frame ,
The method of restoring a resized depth frame to an original depth frame, wherein the number of subpixels of each pixel in the resized depth frame is three.   The method of claim 13, wherein the number of subpixels of each pixel in the original depth frame is three, and the subpixel values of the subpixels are the same.   The method of claim 13, wherein the first pixel and the second pixel in the original depth frame are located in the same row of the original depth frame.   The method of claim 13, wherein the first pixel and the second pixel in the original depth frame are located in the same column of the original depth frame.   The method of claim 13, wherein a size of the original depth frame along a row direction is three times the resized depth frame. An apparatus for restoring a resized depth frame to an original depth frame,
A pixel acquisition unit for acquiring at least one first subpixel value , second subpixel value , and third subpixel value of one pixel in the resized depth frame;
The first subpixel value is stored in the three subpixels of the first pixel in the original depth frame, and the second subpixel value is stored in the three subpixels of the second pixel in the original depth frame. , and a pixel storage unit which Ru is stored within a third three sub-pixels of pixels in the third the original depth frame subpixel values,
The apparatus for restoring a resized depth frame to an original depth frame, wherein the number of sub-pixels of each pixel in the resized depth frame is three.   The apparatus of claim 18, wherein the number of sub-pixels of each pixel in the original depth frame is three, and the sub-pixel value of each pixel is the same.   19. The apparatus of claim 18, wherein the first pixel and the second pixel in the original depth frame are located in the same row of the original depth frame.   The apparatus of claim 18, wherein the first pixel and the second pixel in the original depth frame are located in the same column of the original depth frame.   The apparatus of claim 18, wherein a size of the original depth frame along a row direction is three times the resized depth frame. A system for restoring a resized depth frame to an original depth frame,
A memory unit for storing the resized depth frame;
Obtain at least one first subpixel value , second subpixel value , and third subpixel value of one pixel in the resized depth frame , and use the first subpixel value as the first pixel in the original depth frame. And the second subpixel value is stored in three subpixels of the second pixel in the original depth frame , and the third subpixel value is stored in the third subpixel in the original depth frame. comprising a processing unit Ru is stored in the three sub-pixels of the pixel, and
The system for restoring a resized depth frame to an original depth frame, wherein the number of sub-pixels of each pixel in the resized depth frame is three.   24. The system according to claim 23, wherein the number of subpixels of each pixel in the original depth frame is three, and the subpixel values of the subpixels are the same.   The system of claim 23, wherein the first pixel and the second pixel in the original depth frame are located in the same row of the original depth frame.   The system according to claim 23, wherein the first pixel and the second pixel in the original depth frame are located in the same column of the original depth frame.   The system of claim 23, wherein a size of the original depth frame along a row direction is three times the resized depth frame.