Deep motion and appearance cues for visual tracking
Pattern Recognition Letters(2019)
摘要
•We investigate the impact of deep motion features for visual tracking.•We fuse hand-crafted and deep appearance features with deep motion features in a state-of-the-art tracking framework.•Extensive experiments are performed on the OTB-2015, Temple-Color, and VOT-2015 datasets.•We show that deep motion features lead to significant improvement in tracking performance, leading to state-of-the-artresults.
更多查看译文
关键词
Visual tracking,Deep learning,Optical flow,Discriminative correlation filters
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络