文献类型：期刊文献

英文题名：Technique Report of CVPR 2024 PBDL Challenges

作者：Lu, Xiaoqiang[1]; Jiao, Licheng[1]; Liu, Fang[1]; Liu, Xu[1]; Li, Lingling[1]; Ma, Wenping[1]; Yang, Shuyuan[1]; Xie, Haiyang[2,8]; Zhao, Jian[7,8]; Huang, Shihuang[3]; Cheng, Peng[4]; Shen, Xi[3]; Wang, Zheng[2]; An, Shuai[6]; Zhu, Caizhi[3]; Li, Xuelong[5]; Chen, Linwei[9]; Fu, Ying[9]; Zhang, Tao[10]; Li, Liang[10]; Liu, Yu[11]; Yan, Chenggang[10]; Wang, Zichun[9]; Wang, Qinliang[1]; Gou, Xuejian[1]; Liu, Yang[1]; Yu, Xinyue[12]; Jia, Sen[12]; Zhang, Junpei[12]; Jiao, Licheng[12]; Liu, Xu[12]; Chen, Puhua[12]; Chen, Xiang[13]; Li, Hao[13]; Pan, Jinshan[13]; Xie, Chuanlong[14]; Chen, Hongming[14]; Li, Mingrui[15]; Deng, Tianchen[16]; Huang, Jingwei[17]; Li, Yufeng[14]; Wan, Fei[18,19]; Xu, Bingxin[18,19]; Cheng, Jian[18,19]; Liu, Hongzhe[18,19]; Xu, Cheng[18,19]; Zou, Yuxiang[18,19]; Pan, Weiguo[18,19]; Dai, Songyin[18,19]; Jiang, Linyan[20]; Song, Bingyi[20]; An, Zhuoyu[20]; Lei, Haibo[20]; Luo, Qing[20]; Song, Jie[20]; Liu, Yuan[21]; Li, Qihang[21]; Zhang, Haoyuan[21]; Wang, Lingfeng[21]; Chen, Wei[21]; Luo, Aling[21]; Li, Cheng[22]; Cao, Jun[22]; Chen, Shu[22]; Dou, Zifei[22]; Liu, Xinyu[12]; Zhang, Jing[12]; Zhang, Kexin[12]; Yang, Yuting[12]; Yang, Shuyuan[12]; Zhang, Liwen[23]; Xu, Zhe[23]; Gou, Dingyong[23]; Li, Cong[23]; Xu, Senyan[24]; Zhang, Yunkang[24]; Jiang, Siyuan[24]; Liu, Qinglin[6]; Yu, Wei[6]; Lv, Xiaoqian[6]; Li, Jianing[25]; Zhang, Shengping[6]; Ji, Xiangyang[11]; Zou, Yunhao[9]; Chen, Yuanpei[26]; Zhang, Yuhan[26]; Peng, Weihang[26]; Zhao, Shizhan[12]; Zhang, Yanzhao[12]; Yan, Libo[12]; Lu, Xiaoqiang[12]; Guo, Yuwei[12]; Li, Guoxin[12]; Gao, Qiong[12]; Che, Chenyue[12]; Sun, Long[12]

第一作者：Lu, Xiaoqiang

机构：[1] School of Artificial Intelligence, Xidian University, China; [2] School of Computer Science, Wuhan University, China; [3] Intellindust, China; [4] Beijing Forestry University, China; [5] Institute of AI [TeleAI], China Telecom, China; [6] Harbin Institute of Technology, China; [7] School of Artificial Intelligence, Optics and Electronics [iOPEN], Northwestern Polytechnical University, China; [8] EVOL Lab, Institute of AI [TeleAI], China Telecom, China; [9] Beijing Institute of Technology, China; [10] Lishui Institute of Hangzhou Dianzi University, China; [11] Tsinghua University, China; [12] Intelligent Perception and Image Understanding Lab, Xidian University, China; [13] Nanjing University of Science and Technology, China; [14] Shenyang Aerospace University, China; [15] Dalian University of Technology, China; [16] Shanghai Jiao Tong University, China; [17] University of Electronic Science and Technology, China; [18] Beijing Key Laboratory of Information Service Engineering, Beijing Union University, Beijing, 100101, China; [19] College of Robotics, Beijing Union University, Beijing, 100101, China; [20] Tencent, China; [21] Sanechips Technology Co., LTD., China; [22] Xiaomi inc., China; [23] ZTE Corporation, China; [24] University of Science and Technology, China; [25] Peking University, China; [26] Intelligent Science & Technology Academy, CASIC, China

第一机构：School of Artificial Intelligence, Xidian University, China

年份：2024

外文期刊名：arXiv

收录：EI(收录号：20240281876)

语种：英文

外文关键词：High energy physics

摘要：The intersection of physics-based vision and deep learning presents an exciting frontier for advancing computer vision technologies. By leveraging the principles of physics to inform and enhance deep learning models, we can develop more robust and accurate vision systems. Physics-based vision aims to invert the processes to recover scene properties such as shape, reflectance, light distribution, and medium properties from images. In recent years, deep learning has shown promising improvements for various vision tasks, and when combined with physics-based vision, these approaches can enhance the robustness and accuracy of vision systems. This technical report summarizes the outcomes of the Physics-Based Vision Meets Deep Learning (PBDL) 2024 challenge, held in CVPR 2024 workshop. The challenge consisted of eight tracks, focusing on Low-Light Enhancement and Detection as well as High Dynamic Range (HDR) Imaging. This report details the objectives, methodologies, and results of each track, highlighting the top-performing solutions and their innovative approaches. ? 2024, CC BY.