Mask2Former for Video Instance Segmentation

TL;DR

Mask2Former, originally an image segmentation architecture, achieves state-of-the-art results in video instance segmentation without architectural modifications, demonstrating its versatility and potential as a universal segmentation model.

Contribution

This paper shows that Mask2Former can be directly applied to video segmentation tasks, setting new performance benchmarks without changes to its architecture or training process.

Findings

Achieves 60.4 AP on YouTubeVIS-2019

Achieves 52.6 AP on YouTubeVIS-2021

Demonstrates versatility for video semantic and panoptic segmentation

Abstract

We find Mask2Former also achieves state-of-the-art performance on video instance segmentation without modifying the architecture, the loss or even the training pipeline. In this report, we show universal image segmentation architectures trivially generalize to video segmentation by directly predicting 3D segmentation volumes. Specifically, Mask2Former sets a new state-of-the-art of 60.4 AP on YouTubeVIS-2019 and 52.6 AP on YouTubeVIS-2021. We believe Mask2Former is also capable of handling video semantic and panoptic segmentation, given its versatility in image segmentation. We hope this will make state-of-the-art video segmentation research more accessible and bring more attention to designing universal image and video segmentation architectures.