Name
Abstract | ||
|---|---|---|
We describe novel but simple motion features for the problem of detecting objects in video sequences. Previous approaches either compute optical flow or temporal differences on video frame pairs with various assumptions about stabilization. We describe a combined approach that uses coarse-scale flow and fine-scale temporal difference features. Our approach performs weak motion stabilization by factoring out camera motion and coarse object motion while preserving nonrigid motions that serve as useful cues for recognition. We show results for pedestrian detection and human pose estimation in video sequences, achieving state-of-the-art results in both. In particular, given a fixed detection rate our method achieves a five-fold reduction in false positives over prior art on the Caltech Pedestrian benchmark. Finally, we perform extensive diagnostic experiments to reveal what aspects of our system are crucial for good performance. Proper stabilization, long time-scale features, and proper normalization are all critical. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/CVPR.2013.371 | Computer Vision and Pattern Recognition |
Keywords | Field | DocType |
cameras,feature extraction,image motion analysis,image sequences,object detection,pedestrians,pose estimation,Caltech Pedestrian benchmark,camera motion,coarse object motion,coarse-scale flow difference feature,false positives,fine-scale temporal difference feature,five-fold reduction,human pose estimation,image recognition,motion feature extraction,nonrigid motion preservation,normalization,object detection rate,pedestrian detection,time-scale features,video frame pairs,video sequences,weak motion stabilization | Computer vision,Object detection,Quarter-pixel motion,Pattern recognition,Feature (computer vision),Computer science,Feature extraction,Pose,Artificial intelligence,Motion estimation,Optical flow,Pedestrian detection | Conference |
Volume | Issue | ISSN |
2013 | 1 | 1063-6919 |
Citations | PageRank | References |
52 | 1.92 | 28 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Dennis Park | 1 | 242 | 9.60 |
| C. Lawrence Zitnick | 2 | 7321 | 332.72 |
| deva ramanan | 3 | 10678 | 566.72 |
| Piotr Dollár | 4 | 7999 | 307.07 |