Temporally Distributed Networks for Fast Video Semantic Segmentation

Ping Hu, Fabian Caba, Oliver Wang, Zhe Lin, Stan Sclaroff, Federico Perazzi

Abstract

We present TDNet, a temporally distributed network designed for fast and accurate video semantic segmentation. We observe that features extracted from a certain high-level layer of a deep CNN can be approximated by composing features extracted from several shallower sub-networks. Leveraging the inherent temporal continuity in videos, we distribute these sub-networks over sequential frames. Therefore, at each time step, we only need to perform a lightweight computation to extract a sub-features group from a single sub-network. The full features used for segmentation are then recomposed by application of a novel attention propagation module that compensates for geometry deformation between frames. A grouped knowledge distillation loss is also introduced to further improve the representation power at both full and sub-feature levels. Experiments on Cityscapes, CamVid, and NYUD-v2 demonstrate that our method achieves state-of-the-art accuracy with significantly faster speed and lower latency.
BibTeX
@inproceedings{hu2020temporally,
  author = {Ping Hu and Fabian Caba and Oliver Wang and Zhe Lin and Stan Sclaroff and Federico Perazzi},
  title = {{Temporally Distributed Networks for Fast Video Semantic Segmentation}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year = {2020},
  doi = {10.1109/cvpr42600.2020.00884},
}
Figure