On December 1st, the China Society of Image and Graphics announced the results of the selection for the 2023 Science and Technology Awards. The project "Fundamental Theories and Methods for Large-Scale Efficient Visual Search," jointly completed by Professor Hong Richang, Associate Professor Hu Zhenzhen, and Professor Wang Meng from our institution, along with Professor Zhou Wengang and Professor Li Houqiang from the University of Science and Technology of China, was awarded the second prize in the Natural Science category of the China Society of Image and Graphics.

Established in 1990, the China Society of Image and Graphics is a national-level society approved by the Ministry of Civil Affairs, with independent legal personality. It is a formal group member of the China Association for Science and Technology, consisting of experts, scholars, and related scientific and technical workers engaged in basic theory and applied research in image and graphics, as well as software, hardware technology development, and application promotion. The Science and Technology Award of the China Society of Image and Graphics is set up by the national-level society to stimulate the enthusiasm and creativity of scientific and technical workers in the field of image and graphics in China through incentive mechanisms. It aims to promote technological innovation and industrial development in the field of image and graphics technology. After formal review, preliminary evaluation, public notice, final evaluation, and other procedures, the China Society of Image and Graphics selected nine winners for the 2023 Natural Science Award, including three first prizes and six second prizes.

The theory and methods of large-scale efficient visual search aim to enhance the acquisition and analysis capabilities of visual data resources, fully tapping into the supporting potential of visual data resources and driving innovation in the technology research and development system. To achieve fast and precise retrieval of the required images on the internet, it is necessary to comprehensively construct the search boundary of internet visual data, precisely represent the search targets, and compactly express the search space, establishing a complete visual search ecosystem.

Supported by the National Natural Science Foundation's Outstanding Youth Science Fund, Excellent Youth Science Fund, and General Program, this achievement focuses on feature representation for large-scale efficient visual search. It delves into in-depth research to achieve three goals: completeness of the search boundary, precision of search targets, and compactness of the search space. The project has proposed corresponding theories and methods, resulting in a series of innovative research outcomes. The project's achievements have received positive evaluations and high recognition from academicians in China, the United States, Europe, Australia, and fellows of ACM/IEEE/IAPR. The research outcomes have significantly propelled the development of this field and related areas.