Knowledge Commons of Institute of Automation,CAS
Learn to match: Automatic matching network design for visual tracking | |
Zhang, Zhipeng1,2![]() ![]() ![]() | |
2021 | |
会议名称 | IEEE/CVF International Conference on Computer Vision (ICCV) |
会议日期 | 2021-8 |
会议地点 | Montreal, Canada |
摘要 | Siamese tracking has achieved groundbreaking performance in recent years, where the essence is the efficient matching operator cross correlation and its variants. Besides the remarkable success, it is important to note that the heuristic matching network design relies heavily on expert experience. Moreover, we experimentally find that one sole matching operator is difficult to guarantee stable tracking in all challenging environments. Thus, in this work, we introduce six novel matching operators from the perspective of feature fusion instead of explicit similarity learning, namely Concatenation, Pointwise-Addition, Pairwise-Relation, FiLM, Simple-Transformer and Transductive-Guidance, to explore more feasibility on matching operator selection. The analyses reveal these operators’ selective adaptability on different environment degradation types, which inspires us to combine them to explore complementary features. To this end, we propose binary channel manipulation (BCM) to search for the optimal combination of these operators. BCM determines to retrain or discard one operator by learning its contribution to other tracking steps. By inserting the learned matching networks to a strong baseline tracker Ocean [47], our model achieves favorable gains by 67.2 → 71.4, 52.6 → 58.3, 70.3 → 76.0 success on OTB100, LaSOT, and TrackingNet, respectively. Notably, Our tracker, dubbed AutoMatch, uses less than half of training data/time than the baseline tracker, and runs at 50 FPS using PyTorch. Code and model are released at https://github.com/JudasDie/SOTS. |
收录类别 | EI |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/48528 |
专题 | 多模态人工智能系统全国重点实验室_视频内容安全 |
作者单位 | 1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 2.School of Artificial Intelligence, University of Chinese Academy of Sciences 3.Peng Cheng Laboratory 4.CAS Center for Excellence in Brain Science and Intelligence Technology |
第一作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Zhang, Zhipeng,Liu, Yihao,Wang Xiao,et al. Learn to match: Automatic matching network design for visual tracking[C],2021. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Learn to Match-Autom(1661KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论