Abstract: Video semantic segmentation (VSS) is a computationally expensive task due to the per-frame prediction for videos of high frame rates. In recent work, compact models or adaptive network ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback