Prototypical Cross-Attention Networks for Multiple Object Tracking and Segmentation

TL;DR

Prototypical Cross-Attention Networks (PCAN) leverage spatio-temporal information through prototypes and cross-attention to improve online multiple object tracking and segmentation.

Contribution

The paper introduces PCAN, a novel method that uses space-time prototypes and cross-attention for enhanced object tracking and segmentation.

Findings

Outperforms current state-of-the-art on Youtube-VIS and BDD100K datasets.

Effective for both one-stage and two-stage segmentation frameworks.

Utilizes rich spatio-temporal information for improved accuracy.

Abstract

Multiple object tracking and segmentation requires detecting, tracking, and segmenting objects belonging to a set of given classes. Most approaches only exploit the temporal dimension to address the association problem, while relying on single frame predictions for the segmentation mask itself. We propose Prototypical Cross-Attention Network (PCAN), capable of leveraging rich spatio-temporal information for online multiple object tracking and segmentation. PCAN first distills a space-time memory into a set of prototypes and then employs cross-attention to retrieve rich information from the past frames. To segment each object, PCAN adopts a prototypical appearance module to learn a set of contrastive foreground and background prototypes, which are then propagated over time. Extensive experiments demonstrate that PCAN outperforms current video instance tracking and segmentation competition winners on both Youtube-VIS and BDD100K datasets, and shows efficacy to both one-stage and two-stage segmentation frameworks. Code and video resources are available at http://vis.xyz/pub/pcan.