Papers › EfficientPS: Efficient Panoptic Segmentation

EfficientPS: Efficient Panoptic Segmentation

5 Apr 2020arXiv:2004.02307archive 2025-07-28

Rohit Mohan, Abhinav Valada

Understanding the scene in which an autonomous robot operates is critical for its competent functioning. Such scene comprehension necessitates recognizing instances of traffic participants along with general scene semantics which can be effectively addressed by the panoptic segmentation task. In this paper, we introduce the Efficient Panoptic Segmentation (EfficientPS) architecture that consists of a shared backbone which efficiently encodes and fuses semantically rich multi-scale features. We incorporate a new semantic head that aggregates fine and contextual features coherently and a new variant of Mask R-CNN as the instance head. We also propose a novel panoptic fusion module that congruously integrates the output logits from both the heads of our EfficientPS architecture to yield the final panoptic segmentation output. Additionally, we introduce the KITTI panoptic segmentation dataset that contains panoptic annotations for the popularly challenging KITTI benchmark. Extensive evaluations on Cityscapes, KITTI, Mapillary Vistas and Indian Driving Dataset demonstrate that our proposed architecture consistently sets the new state-of-the-art on all these four benchmarks while being the most efficient and fast panoptic segmentation architecture to date.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

DeepSceneSeg/EfficientPS officialmentioned in papermentioned on GitHubpytorchGPL-3.0 report

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

Instance SegmentationPanoptic SegmentationSegmentationSemantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Panoptic Segmentation Cityscapes test EfficientPS PQ 67.1 #3 of 10 Archive leaderboard report
Panoptic Segmentation Cityscapes test EfficientPS (Cityscapes-fine) PQ 62.9 #7 of 10 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS AP 43.5 #11 of 37 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS PQ 67.5 #11 of 37 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS PQst 70.3 #11 of 37 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS PQth 63.2 #11 of 37 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS mIoU 82.1 #11 of 37 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS (Cityscapes-fine) AP 39.1 #16 of 37 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS (Cityscapes-fine) PQ 64.9 #16 of 37 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS (Cityscapes-fine) PQst 67.7 #16 of 37 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS (Cityscapes-fine) PQth 61.0 #16 of 37 Archive leaderboard report
Panoptic Segmentation Cityscapes val EfficientPS (Cityscapes-fine) mIoU 90.3 #16 of 37 Archive leaderboard report
Panoptic Segmentation Indian Driving Dataset EfficientPS PQ 51.1 #1 of 4 Archive leaderboard report
Panoptic Segmentation KITTI Panoptic Segmentation EfficientPS PQ 43.7 #1 of 4 Archive leaderboard report
Panoptic Segmentation Mapillary val EfficientPS PQ 40.6 #7 of 13 Archive leaderboard report
Semantic Segmentation Cityscapes test EfficientPS Mean IoU (class) 84.21% #10 of 105 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

ConvolutionMask R-CNNRPNRoIAlignSoftmax

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