Deep learning models have a risk of utilizing spurious clues to make
pre...
Motivated by the superior performance of image diffusion models, more an...
Transformer trackers have achieved impressive advancements recently, whe...
Knowledge distillation is widely adopted in semantic segmentation to red...
Multi-person pose estimation and tracking serve as crucial steps for vid...
In this paper, we study the actor-action semantic segmentation problem, ...
Occlusions, complex backgrounds, scale variations and non-uniform
distri...