Abstract | ||
---|---|---|
Learning the spatial-temporal representation of motion information is crucial to human action recognition. Nevertheless, most of the existing features or descriptors cannot capture motion information effectively, especially for long-term motion. To address this problem, this paper proposes a long-term motion descriptor called sequential deep trajectory descriptor (sDTD). Specifically, we project d... |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/TMM.2017.2666540 | IEEE Transactions on Multimedia |
Keywords | DocType | Volume |
Trajectory,Feature extraction,Optical imaging,Cameras,Streaming media,Neural networks,Histograms | Journal | 19 |
Issue | ISSN | Citations |
7 | 1520-9210 | 22 |
PageRank | References | Authors |
1.10 | 0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yemin Shi | 1 | 37 | 9.48 |
Yonghong Tian | 2 | 1057 | 102.81 |
Yaowei Wang | 3 | 134 | 29.62 |
Tiejun Huang | 4 | 1281 | 120.48 |