login in  |  Contact us  |  Join the Society  |  Chinese
About CSIG
current location:Home >About CSIG >Professional Committee
Technical Committee on Traffic Video
Time:2025-12-05      Source: China Society of Image Graphics JGGBA      Share:

1 Full Chinese Name, Full English Name and English Abbreviation of the Technical Committee

Full Chinese Name: 交通视频专业委员会

Full English Name: Technical Committee on Traffic Video

English Abbreviation: TCTV

 

2 Brief Introduction to the Technical Committee

With the continuous advancement of transportation infrastructure construction, China has gradually built a multi-dimensional and comprehensive sensor system in transportation fields such as highways, railways, and urban rail transit. In particular, the traffic visual perception system has provided strong data support for intelligent transportation. In addition, the rapid development of new-generation artificial intelligence technology has also provided powerful algorithmic momentum for the construction of intelligent transportation. Technologies such as automated traffic video analysis, autonomous driving, vehicle-road coordination, and data security are developing rapidly.

 

The Technical Committee on Traffic Video is one of the technical committees under the China Society for Image and Graphics (CSIG), renamed from the Railway Technical Committee of CSIG. The committee is composed of experts, scholars, and relevant scientific and technological workers engaged in traffic-related basic theory and application research, software and hardware technology development, and application promotion in China. Members are from renowned domestic universities, research institutes, and related enterprises. The committee aims to unite and organize professionals in China's traffic-related fields, carry out academic exchange activities, strengthen talent training, promote cooperation between academia and industry, undertake social service work such as knowledge popularization and policy suggestions, and contribute to improving China's scientific research, teaching, application level, and international influence in the field of transportation.

 

3 Information of the Person in Charge of the Technical Committee

Director:

Zhao Yao, Beijing Jiaotong University

Deputy Directors:

Tian Yonghong, Peking University Shenzhen Graduate School/Pengcheng Laboratory

Lü Ke, University of Chinese Academy of Sciences

Yang Jun, Lanzhou Jiaotong University

Shi Tianyun, China Academy of Railway Sciences Corporation Limited

Secretary-General:

Li Jia, Beihang University

 

4 Contact Person and Contact Information of the Technical Committee

Contact Person: Li Jia

Contact Email: jiali@buaa.edu.cn

 

5 Representative Achievements

1) High-Efficiency Visual Analysis Technology

Core Researchers: Wei Shikui, Lin Chunyu, Zhu Zhenfeng, Hu Shaohai

Focusing on the widely existing massive visual information, special technical research has been carried out, breaking through a series of key technologies such as high-efficiency visual representation and full-scene weak and small target detection and recognition. The technology has been applied and promoted in multiple fields, significantly improving the application efficiency in specific scenarios. Relevant technologies have been successfully applied in various equipment in civil and national defense fields, achieving important civil and national defense economic benefits.

 

2) Graph Transformer-Based Question Answering Model for Financial Scenarios

Core Researchers: Cao Meng, Jiang Ji, Ye Qichen, Zou Yuexian

Financial Community Question Answering (FCQA) is a new and challenging task introduced at the World Artificial Intelligence Conference (WAIC 2022). FCQA aims to respond to users' queries in the financial field under market fluctuation conditions and provide reasonable answers with the help of heterogeneous knowledge sources. We propose a Graph Transformer-based model to solve this problem to achieve efficient multi-source information fusion. By encoding heterogeneous input information sources into different nodes, we explicitly model their relevance to queries and answers. In addition, we update each edge in a hierarchical manner to transmit accurate information sources. We won the first place among 4,278 participating teams and achieved a BLUE score 5.07 times higher than the second place.


Copyright © 2025 China Society of Image Graphics JGGBA No. 11010802035643 ICP No. 12009057-1

Address:No. 95, Zhongguancun East Road,Haidian District, Beijing Postcode: 100190

Technical support:zkf