[ PAPER ] · 2021 · Computer Vision and Pattern Recognition
Masked-attention Mask Transformer for Universal Image Segmentation
Bowen Cheng, Ishan Misra, A. Schwing, Alexander Kirillov, Rohit Girdhar
[ TLDR ]
Mask2Former is presented, a new archi-tecture capable of addressing any image segmentation task (panoptic, instance or semantic), and sets a new state-of-the-art for panoptic segmentation, instance segmentation and semantic segmentation.
[ ABSTRACT ]
Image segmentation groups pixels with different semantics, e.g., category or instance membership. Each choice of semantics defines a task. While only the semantics of each task differ, current research focuses on designing spe-cialized architectures for each task. We present Masked- attention Mask Transformer (Mask2Former), a new archi-tecture capable of addressing any image segmentation task (panoptic, instance or semantic). Its key components in-clude masked attention, which extracts localized features by constraining cross-attention within predicted mask regions. In addition to reducing the research effort by at least three times, it outperforms the best specialized architectures by a significant margin on four popular datasets. Most no-tably, Mask2Former sets a new state-of-the-art for panoptic segmentation (57.8 PQ on COCO), instance segmentation (50.1 AP on COCO) and semantic segmentation (57.7 mIoU onADE20K).