Papers › MaX-DeepLab: End-to-End Panoptic Segmentation with Mask Transformers
MaX-DeepLab: End-to-End Panoptic Segmentation with Mask Transformers
Huiyu Wang, Yukun Zhu, Hartwig Adam, Alan Yuille, Liang-Chieh Chen
We present MaX-DeepLab, the first end-to-end model for panoptic segmentation. Our approach simplifies the current pipeline that depends heavily on surrogate sub-tasks and hand-designed components, such as box detection, non-maximum suppression, thing-stuff merging, etc. Although these sub-tasks are tackled by area experts, they fail to comprehensively solve the target task. By contrast, our MaX-DeepLab directly predicts class-labeled masks with a mask transformer, and is trained with a panoptic quality inspired loss via bipartite matching. Our mask transformer employs a dual-path architecture that introduces a global memory path in addition to a CNN path, allowing direct communication with any CNN layers. As a result, MaX-DeepLab shows a significant 7.1% PQ gain in the box-free regime on the challenging COCO dataset, closing the gap between box-based and box-free methods for the first time. A small variant of MaX-DeepLab improves 3.0% PQ over DETR with similar parameters and M-Adds. Furthermore, MaX-DeepLab, without test time augmentation, achieves new state-of-the-art 51.3% PQ on COCO test-dev set. Code is available at https://github.com/google-research/deeplab2.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Panoptic Segmentation | COCO minival | MaX-DeepLab-L (single-scale) | PQ | 51.1 | #22 of 31 | Archive leaderboard | report |
| Panoptic Segmentation | COCO minival | MaX-DeepLab-L (single-scale) | PQst | 42.2 | #22 of 31 | Archive leaderboard | report |
| Panoptic Segmentation | COCO minival | MaX-DeepLab-L (single-scale) | PQth | 57.0 | #22 of 31 | Archive leaderboard | report |
| Panoptic Segmentation | COCO test-dev | MaX-DeepLab-L (single-scale) | PQ | 51.3 | #12 of 38 | Archive leaderboard | report |
| Panoptic Segmentation | COCO test-dev | MaX-DeepLab-L (single-scale) | PQst | 42.4 | #12 of 38 | Archive leaderboard | report |
| Panoptic Segmentation | COCO test-dev | MaX-DeepLab-L (single-scale) | PQth | 57.2 | #12 of 38 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections