maxtron-video-panoptic-segmentation.github.io - MaXTron: Mask Transformer with Trajectory Attention for Video Panoptic Segmentation

home > domains > maxtron-video-panoptic-segmentation.github.io

Description: MaXTron: Mask Transformer with Trajectory Attention for Video Panoptic Segmentation

video panoptic segmentation (1) video instance segmentation (1) trajectory attention (1)

Example domain paragraphs

MaXTron is a simple yet effective unified meta-architecture for video segmentation.

It enriches existing clip-level segmenters by improving both the within-clip and cross-clip tracking ability. Abstract Video panoptic segmentation requires consistently segmenting (for both 'thing' and 'stuff' classes) and tracking objects in a video over time. In this work, we present MaXTron, a general framework that exploits Mask XFormer with Trajectory Attention to tackle the task. MaXTron enriches an off-the-shelf mask transformer by leveraging trajectory attention. The deployed mask transformer takes

Links to maxtron-video-panoptic-segmentation.github.io (1)

• tacju.github.io Ju He