Body part (BP) classification is important in human image analysis. In this paper, we propose an optical flow based pixel-wise BP classifier using random forest (RF) in a monocular video. Running the BP classifier on each optical frame generates noisy mis classifications. We integrate the classifier with a human detector and temporal voting to reduce most of misclassified pixels. The proposed method is evaluated quantitatively on a real world dataset and showed better performance.
Publications
International Conference
Human Body Part Classification from Optical Flow
조회 수 459
댓글 0
저 자 | Junsik Kim, Kyungdon Joo, Tae-Hyun Oh, In So Kweon |
---|---|
학 회 | The International Conference on Ubiquitous Robots and Ambient Intelligence (URAI) |
논문일시(Year) | 2016 |
논문일시(Month) | 08 |
Prev All-around Depth from Small Motion with A Spherical Panoramic...
All-around Depth from Small Motion with A Spherical Panoramic...
2016.07.14by
Asymmetric Stereo with Catadioptric Lens: High Quality Image Generation for Intelligent Robot Next
Asymmetric Stereo with Catadioptric Lens: High Quality Image Generation for Intelligent Robot
2016.07.05by 조동현