PanSR: An Object-Centric Mask Transformer for Panoptic Segmentation

TL;DR

PanSR introduces an object-centric mask transformer that improves small object detection and scene segmentation in crowded scenes, achieving state-of-the-art results on multiple benchmarks.

Contribution

It proposes a novel panoptic segmentation method that addresses key shortcomings of existing mask-transformer approaches, notably enhancing small object detection and reducing instance merging.

Findings

+3.4 PQ improvement on LaRS benchmark

State-of-the-art performance on Cityscapes

Effective mitigation of instance merging and small-object detection

Abstract

Panoptic segmentation is a fundamental task in computer vision and a crucial component for perception in autonomous vehicles. Recent mask-transformer-based methods achieve impressive performance on standard benchmarks but face significant challenges with small objects, crowded scenes and scenes exhibiting a wide range of object scales. We identify several fundamental shortcomings of the current approaches: (i) the query proposal generation process is biased towards larger objects, resulting in missed smaller objects, (ii) initially well-localized queries may drift to other objects, resulting in missed detections, (iii) spatially well-separated instances may be merged into a single mask causing inconsistent and false scene interpretations. To address these issues, we rethink the individual components of the network and its supervision, and propose a novel method for panoptic segmentation PanSR. PanSR effectively mitigates instance merging, enhances small-object detection and increases performance in crowded scenes, delivering a notable +3.4 PQ improvement over state-of-the-art on the challenging LaRS benchmark, while reaching state-of-the-art performance on Cityscapes. The code and models will be publicly available at https://github.com/lojzezust/PanSR.