Search Results

Weakly supervised object localization (WSOL) strives to localize objects with only image-level supervision. WSOL often faces challenges such as incomplete localization due to classifier bias and over-localization in real scenes where objects and ...

• Present a novel Semantic-Constraint Matching(SeCM) plug-in module under ViT.
• Introduce a semantic matching(SM) module to mitigate divergent activation.
• Address the bias issue by presenting a multi-modal cross-attention(MCA) module.

Fine-tuning is an effective technique to enhance network performance in scenarios with limited labeled data. To achieve this, recent methods exploit the knowledge mined in the source model (e.g., feature maps) to construct an extra regularization ...

The location technology of information sources in social networks is a major factor in exploring the means of information propagation. In this context, most existing methods ignore the direction of infected nodes and fail to make full use of the ...

The whole herb of Euphorbia helioscopia has been traditionally used for treating pulmonary tuberculosis, malaria, warts, lung cancer and bacillary dysentery for a long time in China. However, E. helioscopia seeds are often discarded ...

• Phytochemistry of EHS was identified by widely targeted metabolomics.
• Anti-Pulmonary fibrosis mechanism of EHS was identified by network pharmacology.
• EHS is proved as industrially medicinal resource.

Distant supervision (DS) can automatically generate annotated data for relation extraction (RE) with knowledge bases and corpora. The existing DS methods that train on bags selected by attention mechanism are susceptible to noisy bags ...

This paper studies a new, practical but challenging problem, called Class-Incremental Unsupervised Domain Adaptation (CI-UDA), where the labeled source domain contains all classes, but the classes in the unlabeled target domain increase ...

The network structure exhibits a variety of changes over time. Fusing this structure and the development of communities in dynamic networks plays an important role in analyzing the evolution and development of the entire network. How ...

In the modern world, English education institutions are more concerned about how to recommend the required resources to all types of learners and how to improve the application value of the resources. Therefore, this study introduces and combines the ...

Named entity recognition (NER) is crucial in various natural language processing (NLP) tasks. However, the nested entities which are common in practical corpus are often ignored in most of current NER models. To extract the nested ...

Retinal image segmentation plays an important role in automatic disease diagnosis. This task is very challenging because the complex structure and texture information are mixed in a retinal image, and distinguishing the information is difficult. ...

Path planning is a task of primary importance for unmanned ship, but current algorithms are complex and inefficient. In this paper, we propose a Rapidly-Exploring Random Tree algorithm (RRT) for path planning of unmanned ship, which can obtain an ...

Feature extraction and classification algorithm is important to determine the accuracy of classification. The terrain recognition of a legged robot has higher requirements on real-time classification. Considering the traditional training methods is ...

According to the data characteristics of the assembled-military container failure cases, a new method, named as data mining was put forward in this paper. The necessity of data mining method was introduced in failure analysis domain. The basic principle ...

In Line with the promotion of Zheng-wheat 7698 with strong gluten, this article analyzed the growth of Zheng-wheat 7698 and the relevant super-high yield cultivation technique according to the researChes in the field and experiments in small areas. The ...

An expert fuzzy-PID fusion type controller is designed for VAV air conditioning system base on plenty of experience of practical projects. The principle of a typical VAV air conditioning system is introduced firstly. And the control characteristic ...

An expert fuzzy-PID fusion type controller is designed for VAV air conditioning system base on plenty of experience of practical projects. The principle of a typical VAV air conditioning system is introduced firstly. And the control characteristic ...

A novel image retrieval of JPEG image is proposed combing the statistical distribution with spatial information of the DCT coefficients. Firstly, according to the distribution of DC coefficients, a robust DC feature is presented. Meantime, several AC ...

With the personal request for treatment and the pressure of decreasing accident, it is necessary particularly to use a new technology in hospital. Using ZigBee wireless network, the information of patient will be send to doctor more faster than ever. ...

A new system design aggress with the CAN2.0B is put forward in the paper. It focus on the hardware and software design of intelligent node at the core of microprocessor S3C44B0, which is a SAMSUNG's 32-bit RISC high performance micro-controller. MCP2510 ...

The application of wireless sensor mesh networks in Intelligent Transportation System (ITS)is presented in this paper. For a short distance transmission, the data collector collects data from Reduced Function Device (RFD) sensors node using an efficient ...