Channel-wise Motion Features for Efficient Motion Segmentation
This addresses the need for efficient, real-time motion segmentation in safety-critical robotics applications like autonomous driving, representing an incremental improvement over existing methods.
The paper tackles the problem of high computational cost in motion segmentation models for autonomous driving by proposing Channel-wise Motion Features, which achieve about 4 times the FPS of state-of-the-art models while maintaining equivalent accuracy and reducing parameters to about 25%.
For safety-critical robotics applications such as autonomous driving, it is important to detect all required objects accurately in real-time. Motion segmentation offers a solution by identifying dynamic objects from the scene in a class-agnostic manner. Recently, various motion segmentation models have been proposed, most of which jointly use subnetworks to estimate Depth, Pose, Optical Flow, and Scene Flow. As a result, the overall computational cost of the model increases, hindering real-time performance. In this paper, we propose a novel cost-volume-based motion feature representation, Channel-wise Motion Features. By extracting depth features of each instance in the feature map and capturing the scene's 3D motion information, it offers enhanced efficiency. The only subnetwork used to build Channel-wise Motion Features is the Pose Network, and no others are required. Our method not only achieves about 4 times the FPS of state-of-the-art models in the KITTI Dataset and Cityscapes of the VCAS-Motion Dataset, but also demonstrates equivalent accuracy while reducing the parameters to about 25$\%$.