English

Local Memory Attention for Fast Video Semantic Segmentation

Computer Vision and Pattern Recognition 2021-09-28 v2

Abstract

We propose a novel neural network module that transforms an existing single-frame semantic segmentation model into a video semantic segmentation pipeline. In contrast to prior works, we strive towards a simple, fast, and general module that can be integrated into virtually any single-frame architecture. Our approach aggregates a rich representation of the semantic information in past frames into a memory module. Information stored in the memory is then accessed through an attention mechanism. In contrast to previous memory-based approaches, we propose a fast local attention layer, providing temporal appearance cues in the local region of prior frames. We further fuse these cues with an encoding of the current frame through a second attention-based module. The segmentation decoder processes the fused representation to predict the final semantic segmentation. We integrate our approach into two popular semantic segmentation networks: ERFNet and PSPNet. We observe an improvement in segmentation performance on Cityscapes by 1.7% and 2.1% in mIoU respectively, while increasing inference time of ERFNet by only 1.5ms.

Keywords

Cite

@article{arxiv.2101.01715,
  title  = {Local Memory Attention for Fast Video Semantic Segmentation},
  author = {Matthieu Paul and Martin Danelljan and Luc Van Gool and Radu Timofte},
  journal= {arXiv preprint arXiv:2101.01715},
  year   = {2021}
}

Comments

Accepted at IROS 2021. Source code is available at https://github.com/mattpfr/lmanet

R2 v1 2026-06-23T21:48:48.498Z