Point Cloud Densification for 3D Gaussian Splatting from Sparse Input Views
Pages 896 - 904
Abstract
The technique of 3D Gaussian splatting (3DGS) has demonstrated its effectiveness and efficiency in rendering photo-realistic images for novel view synthesis. However, 3DGS requires a high density of camera coverage, and its performance inevitably degrades with sparse training views, which significantly limits its applicability in real-world scenarios. In recent years, many researchers have explored the use of depth information to alleviate this problem, but the performance of their methods is sensitive to the accuracy of depth estimation. To this end, we propose an efficient method to enhance the performance of 3DGS with sparse training views. Specifically, instead of applying depth maps for regularization, we propose a densification method that generates high-quality point clouds, providing a superior initialization for 3D Gaussians. Furthermore, we propose Systematically Angle of View Sampling (SAOVS), which employs Spherical Linear Interpolation (SLERP) and linear interpolation for side view sampling, to determine unseen views outside the training data for semantic pseudo-label regularization. Experiments show that our proposed method significantly outperforms other leading 3D rendering models on the ScanNet dataset and the LLFF dataset. In particular, compared with the conventional 3DGS method, our proposed method achieves performance gains of up to 1.71dB in PSNR and 0.07 in SSIM. In addition, the novel view synthesis produced by our method demonstrates the highest visual quality with minimal distortions.
Supplemental Material
MP4 File - 4055-video.mp4
Novel view synthesis has gained increasing importance due to the widespread availability of VR devices. Previous methods for sparse input data introduced various regularization techniques to the optimization process but encountered significant challenges. In our approach, we propose a point cloud densification method. Specifically, we use estimated depth information to inversely unproject RGB images into 3D world space, resulting in a dense point cloud. This dense point cloud is then used to initialize the 3DGS model. We introduce semantic constraints into the optimization process to ensure that 3D objects maintain consistent structure and detail across different views. To support this, we propose systematically Angle of View Sampling for generating coherent side views. Our method synthesizes images with fewer artifacts while preserving rich, detailed textures. Notably, it produces sharper object appearances and achieves superior performance across various evaluation metrics.
- Download
- 13.06 MB
References
[1]
Reiner Birkl, Diana Wofk, and Matthias Müller. 2023. MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimation. arXiv preprint arXiv:2307.14460 (2023).
[2]
Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. 2021. Emerging properties in self-supervised vision transformers. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 9650--9660.
[3]
Anirban Chakraborty, Manaar Alam, Vishal Dey, Anupam Chattopadhyay, and Debdeep Mukhopadhyay. 2018. Adversarial attacks and defences: A survey. arXiv preprint arXiv:1810.00069 (2018).
[4]
Anpei Chen, Zexiang Xu, Fuqiang Zhao, Xiaoshuai Zhang, Fanbo Xiang, Jingyi Yu, and Hao Su. 2021. Mvsnerf: Fast generalizable radiance field reconstruction from multi-view stereo. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 14124--14133.
[5]
Guikun Chen and Wenguan Wang. 2024. A survey on 3d gaussian splatting. arXiv preprint arXiv:2401.03890 (2024).
[6]
Jaeyoung Chung, Suyoung Lee, Hyeongjin Nam, Jaerin Lee, and Kyoung Mu Lee. 2023. Luciddreamer: Domain-free generation of 3d gaussian splatting scenes. arXiv preprint arXiv:2311.13384 (2023).
[7]
Jaeyoung Chung, Jeongtaek Oh, and Kyoung Mu Lee. 2023. Depth-regularized optimization for 3d gaussian splatting in few-shot images. arXiv preprint arXiv:2311.13398 (2023).
[8]
Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, and Matthias Nießner. 2017. ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
[9]
Kangle Deng, Andrew Liu, Jun-Yan Zhu, and Deva Ramanan. 2022. Depthsupervised nerf: Fewer views and faster training for free. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 12882-- 12891.
[10]
Shirin Nasr Esfahani, Shahram Latifi, et al. 2019. ASurvey OF STATE-OF-THEART GAN-BASED APPROACHES TO IMAGE SYNTHESIS. In Computer Science & Information Technology (CS & IT) Computer Science Conference Proceedings (CSCP). 63--76.
[11]
Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative adversarial nets. Advances in Neural Information Processing Systems (NeurIPS) 27 (2014).
[12]
Tao Hu, Shu Liu, Yilun Chen, Tiancheng Shen, and Jiaya Jia. 2022. Efficientnerf efficient neural radiance fields. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 12902--12911.
[13]
Ajay Jain, Matthew Tancik, and Pieter Abbeel. 2021. Putting nerf on a diet: Semantically consistent few-shot view synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 5885--5894.
[14]
Bernhard Kerbl, Georgios Kopanas, Thomas Leimkühler, and George Drettakis. 2023. 3D Gaussian Splatting for Real-Time Radiance Field Rendering. ACM Transactions on Graphics 42, 4 (2023).
[15]
Xinhai Li, Huaibin Wang, and Kuo-Kun Tseng. 2023. Gaussiandiffusion: 3d gaussian splatting for denoising diffusion probabilistic models with structured noise. arXiv preprint arXiv:2311.11221 (2023).
[16]
Jae Hyun Lim and Jong Chul Ye. 2017. Geometric gan. arXiv preprint arXiv:1705.02894 (2017).
[17]
Jonathon Luiten, Georgios Kopanas, Bastian Leibe, and Deva Ramanan. 2023. Dynamic 3d gaussians: Tracking by persistent dynamic view synthesis. arXiv preprint arXiv:2308.09713 (2023).
[18]
Ben Mildenhall, Pratul P. Srinivasan, Rodrigo Ortiz-Cayon, Nima Khademi Kalantari, Ravi Ramamoorthi, Ren Ng, and Abhishek Kar. 2019. Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelines. ACM Transactions on Graphics (2019).
[19]
Ben Mildenhall, Pratul P Srinivasan, Rodrigo Ortiz-Cayon, Nima Khademi Kalantari, Ravi Ramamoorthi, Ren Ng, and Abhishek Kar. 2019. Local light field fusion: Practical view synthesis with prescriptive sampling guidelines. ACM Transactions on Graphics 38, 4 (2019), 1--14.
[20]
Ben Mildenhall, Pratul P Srinivasan, Matthew Tancik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. 2021. Nerf: Representing scenes as neural radiance fields for view synthesis. Commun. ACM 65, 1 (2021), 99--106.
[21]
Michael Niemeyer, Jonathan T Barron, Ben Mildenhall, Mehdi SM Sajjadi, Andreas Geiger, and Noha Radwan. 2022. Regnerf: Regularizing neural radiance fields for view synthesis from sparse inputs. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 5480--5490.
[22]
Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. 2021. Learning transferable visual models from natural language supervision. In Proceedings of the International Conference on Machine Learning (ICML). PMLR, 8748--8763.
[23]
René Ranftl, Alexey Bochkovskiy, and Vladlen Koltun. 2021. Vision Transformers for Dense Prediction. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) (2021).
[24]
René Ranftl, Katrin Lasinger, David Hafner, Konrad Schindler, and Vladlen Koltun. 2022. Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero- Shot Cross-Dataset Transfer. IEEE Transactions on Pattern Analysis and Machine Intelligence 44, 3 (2022).
[25]
Barbara Roessle, Jonathan T Barron, Ben Mildenhall, Pratul P Srinivasan, and Matthias Nießner. 2022. Dense depth priors for neural radiance fields from sparse input views. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 12892--12901.
[26]
Divya Saxena and Jiannong Cao. 2021. Generative adversarial networks (GANs) challenges, solutions, and future directions. ACM Computing Surveys (CSUR) 54, 3 (2021), 1--42.
[27]
Johannes Lutz Schönberger and Jan-Michael Frahm. 2016. Structure-from-Motion Revisited. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
[28]
Johannes Lutz Schönberger, True Price, Torsten Sattler, Jan-Michael Frahm, and Marc Pollefeys. 2016. A Vote-and-Verify Strategy for Fast Spatial Verification in Image Retrieval. In Proceedings of the Asian Conference on Computer Vision (ACCV).
[29]
Johannes Lutz Schönberger, Enliang Zheng, Marc Pollefeys, and Jan-Michael Frahm. 2016. Pixelwise View Selection for Unstructured Multi-View Stereo. In Proceedings of the European Conference on Computer Vision (ECCV).
[30]
Ken Shoemake. 1985. Animating rotation with quaternion curves. In Proceedings of the 12th annual conference on Computer graphics and interactive techniques. 245--254.
[31]
Nagabhushan Somraj and Rajiv Soundararajan. 2023. ViP-NeRF: Visibility Prior for Sparse Input Neural Radiance Fields. In ACM Special Interest Group on Computer Graphics and Interactive Techniques (SIGGRAPH). https://doi.org/10.1145/ 3588432.3591539
[32]
GuangcongWang, Zhaoxi Chen, Chen Change Loy, and Ziwei Liu. 2023. Sparsenerf: Distilling depth ranking for few-shot novel view synthesis. arXiv preprint arXiv:2303.16196 (2023).
[33]
Zhou Wang, Alan C Bovik, Hamid R Sheikh, and Eero P Simoncelli. 2004. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing 13, 4 (2004), 600--612.
[34]
Guanjun Wu, Taoran Yi, Jiemin Fang, Lingxi Xie, Xiaopeng Zhang, Wei Wei, Wenyu Liu, Qi Tian, and XinggangWang. 2023. 4d gaussian splatting for real-time dynamic scene rendering. arXiv preprint arXiv:2310.08528 (2023).
[35]
Haolin Xiong, Sairisheek Muttukuru, Rishi Upadhyay, Pradyumna Chari, and Achuta Kadambi. 2023. Sparsegs: Real-time 360° sparse view synthesis using gaussian splatting. arXiv preprint arXiv:2312.00206 (2023).
[36]
Dejia Xu, Yifan Jiang, PeihaoWang, Zhiwen Fan, Humphrey Shi, and Zhangyang Wang. 2022. Sinnerf: Training neural radiance fields on complex scenes from a single image. In Proceedings of the European Conference on Computer Vision (ECCV). Springer, 736--753.
[37]
Taoran Yi, Jiemin Fang, Guanjun Wu, Lingxi Xie, Xiaopeng Zhang, Wenyu Liu, Qi Tian, and Xinggang Wang. 2023. Gaussiandreamer: Fast generation from text to 3d gaussian splatting with point cloud priors. arXiv preprint arXiv:2310.08529 (2023).
[38]
Alex Yu, Vickie Ye, Matthew Tancik, and Angjoo Kanazawa. 2021. pixelnerf: Neural radiance fields from one or few images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 4578--4587.
[39]
Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. 2018. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 586--595.
[40]
Shengyu Zhao, Zhijian Liu, Ji Lin, Jun-Yan Zhu, and Song Han. 2020. Differentiable augmentation for data-efficient gan training. Advances in Neural Information Processing Systems (NeurIPS) 33 (2020), 7559--7570.
[41]
Zehao Zhu, Zhiwen Fan, Yifan Jiang, and Zhangyang Wang. 2023. FSGS: Real-Time Few-shot View Synthesis using Gaussian Splatting. arXiv preprint arXiv:2312.00451 (2023).
[42]
Matthias Zwicker, Hanspeter Pfister, Jeroen Van Baar, and Markus Gross. 2001. EWA volume splatting. In Proceedings Visualization, 2001. VIS'01. IEEE, 29--538.
Index Terms
- Point Cloud Densification for 3D Gaussian Splatting from Sparse Input Views
Recommendations
AbsGS: Recovering Fine Details in 3D Gaussian Splatting
MM '24: Proceedings of the 32nd ACM International Conference on Multimedia3D Gaussian Splatting (3D-GS) technique couples 3D Gaussian primitives with differentiable rasterization to achieve high-quality novel view synthesis results while providing advanced real-time rendering performance. However, due to the flaw of its ...
Relightable 3D Gaussians: Realistic Point Cloud Relighting with BRDF Decomposition and Ray Tracing
Computer Vision – ECCV 2024AbstractIn this paper, we present a novel differentiable point-based rendering framework to achieve photo-realistic relighting. To make the reconstructed scene relightable, we enhance vanilla 3D Gaussians by associating extra properties, including normal ...
GS3: Efficient Relighting with Triple Gaussian Splatting
SA '24: SIGGRAPH Asia 2024 Conference PapersWe present a spatial and angular Gaussian based representation and a triple splatting process, for real-time, high-quality novel lighting-and-view synthesis from multi-view point-lit input images. To describe complex appearance, we employ a Lambertian ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
October 2024
11719 pages
ISBN:9798400706868
DOI:10.1145/3664647
- General Chairs:
- Jianfei Cai,
- Mohan Kankanhalli,
- Balakrishnan Prabhakaran,
- Susanne Boll,
- Program Chairs:
- Ramanathan Subramanian,
- Liang Zheng,
- Vivek K. Singh,
- Pablo Cesar,
- Lexing Xie,
- Dong Xu
Copyright © 2024 Owner/Author.
This work is licensed under a Creative Commons Attribution International 4.0 License.
Sponsors
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 28 October 2024
Check for updates
Author Tags
Qualifiers
- Research-article
Conference
MM '24
Sponsor:
MM '24: The 32nd ACM International Conference on Multimedia
October 28 - November 1, 2024
Melbourne VIC, Australia
Acceptance Rates
MM '24 Paper Acceptance Rate 1,150 of 4,385 submissions, 26%;
Overall Acceptance Rate 2,145 of 8,556 submissions, 25%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 260Total Downloads
- Downloads (Last 12 months)260
- Downloads (Last 6 weeks)155
Reflects downloads up to 09 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in