Multi-stream spatiotemporal networks for driver fatigue detection from infrared and depth videos
This thesis presents new methods for incorporating multi-stream networks into the driver fatigue detection system. For the depth video-based method, a two-stream CNN architecture is proposed to incorporate spatial information of the current depth frame and temporal information of neighboring depth f...
Saved in:
| 主要作者: | |
|---|---|
| 其他作者: | |
| 格式: | Thesis-Doctor of Philosophy |
| 語言: | English |
| 出版: |
Nanyang Technological University
2023
|
| 主題: | |
| 在線閱讀: | https://hdl.handle.net/10356/164796 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 機構: | Nanyang Technological University |
| 語言: | English |
| 總結: | This thesis presents new methods for incorporating multi-stream networks into the driver fatigue detection system. For the depth video-based method, a two-stream CNN architecture is proposed to incorporate spatial information of the current depth frame and temporal information of neighboring depth frames which is represented by motion vectors. For the infrared video-based method, a convolutional three-stream network is proposed to incorporate current-infrared-frame-based spatial information, optical-flow-based short-term temporal information of two consecutive infrared frames, and optical flow-motion history image-based temporal information within the infrared video sequence. |
|---|