71 citations
- Huazhong University of Science and TechnologyCN8 papers
- Chinese Academy of SciencesCN7 papers
- Institute of AutomationCN7 papers
- Nanjing UniversityCN7 papers
- Tsinghua UniversityCN7 papers
- University of Chinese Academy of SciencesCN6 papers
- Wuhan UniversityCN5 papers
- Institute of AcousticsCN4 papers
- Center for Excellence in Brain Science and Intelligence TechnologyCN3 papers
- Chinese University of Hong KongHK2 papers
- Google (United States)US2 papers
- Horizon Research (United States)US2 papers
11 papers · 1 filter
Diversity Transfer Network for Few-Shot Learning
Mengting Chen, Yuxin Fang, Xinggang Wang +6
Few-shot learning is a challenging task that aims at training a classifier for unseen classes with only a few training examples. The main difficulty of few-shot learning lies in th…
Learned Video Compression via Joint Spatial-Temporal Correlation Exploration
Haojie Liu, Han shen, Lichao Huang +3
Traditional video compression technologies have been developed over decades in pursuit of higher coding efficiency. Efficient temporal information representation plays a key role i…
AugFPN: Improving Multi-scale Feature Learning for Object Detection
Chaoxu Guo, Bin Fan, Qian Zhang +2
Current state-of-the-art detectors typically exploit feature pyramid to detect objects at different scales. Among them, FPN is one of the representative works that build a feature…
Siamese Natural Language Tracker: Tracking by Natural Language Descriptions with Siamese Trackers
Qi Feng, Vitaly Ablavsky, Qinxun Bai +1
We propose a novel Siamese Natural Language Tracker (SNLT), which brings the advancements in visual tracking to the tracking by natural language (NL) descriptions task. The propose…
Learning Where to Focus for Efficient Video Object Detection
Zhengkai Jiang, Yu Liu, Ceyuan Yang +5
Transferring existing image-based detectors to the video is non-trivial since the quality of frames is always deteriorated by part occlusion, rare pose, and motion blur. Previous a…
Learning Enhanced Resolution-wise features for Human Pose Estimation
Kun Zhang, Peng He, Ping Yao +6
Recently, multi-resolution networks (such as Hourglass, CPN, HRNet, etc.) have achieved significant performance on pose estimation by combining feature maps of various resolutions.…