Browse State-of-the-Art › Panoptic Segmentation

Panoptic Segmentation

257 papers with code · 27 benchmarks · 37 datasets archive 2025-07-28

Computer Vision

Panoptic Segmentation is a computer vision task that combines semantic segmentation and instance segmentation to provide a comprehensive understanding of the scene. The goal of panoptic segmentation is to segment the image into semantically meaningful parts or regions, while also detecting and distinguishing individual instances of objects within those regions. In a given image, every pixel is assigned a semantic label, and pixels belonging to "things" classes (countable objects with instances, like cars and people) are assigned unique instance IDs. ( Image credit: Detectron2 )

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

27 leaderboard tables shown for this task, 27 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 27 until expanded.

Syntology 1 newer paper was judged to report on this task's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 0 placed): PanNuke (1 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
COCO test-dev (38 rows) Mask DINO (single scale) Mask DINO: Towards A Unified Transformer-based Framework for... code Syntology community repositories only · 11 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 1 violated, 1 with no contract checked; 9 where Syntology's instrument failed) · 2 unverified Compare
Cityscapes val (37 rows) ViT-P (OneFormer, InternImage-H) The Missing Point in Vision Transformers for Universal Image Segmentation code — Compare
COCO minival (31 rows) HyperSeg (Swin-B) HyperSeg: Towards Universal Visual Segmentation with Large Language Model code Syntology official (archive's flag): 13 ran · 13 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 4 honoured, 2 violated, 2 with no contract checked; 5 where Syntology's instrument failed) · 4 unverified Compare
ADE20K val (25 rows) OneFormer (InternImage-H, emb_dim=256, single-scale, 896x896) OneFormer: One Transformer to Rule Universal Image Segmentation code Syntology official (archive's flag): 4 ran · 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified Compare
Mapillary val (13 rows) OneFormer (DiNAT-L, single-scale) OneFormer: One Transformer to Rule Universal Image Segmentation code Syntology official (archive's flag): 4 ran · 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified Compare
Cityscapes test (10 rows) OneFormer (ConvNeXt-L, single-scale, Mapillary Vistas-Pretrained) OneFormer: One Transformer to Rule Universal Image Segmentation code Syntology official (archive's flag): 4 ran · 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified Compare
LaRS (8 rows) Mask2Former (Swin-B) LaRS: A Diverse Panoptic Maritime Obstacle Detection Dataset and Benchmark code — Compare
S3DIS Area5 (5 rows) SuperCluster Scalable 3D Panoptic Segmentation As Superpoint Graph Clustering code — Compare
ScanNetV2 (5 rows) OneFormer3D OneFormer3D: One Transformer for Unified Point Cloud Segmentation code — Compare
Indian Driving Dataset (4 rows) EfficientPS EfficientPS: Efficient Panoptic Segmentation code — Compare
KITTI Panoptic Segmentation (4 rows) EfficientPS EfficientPS: Efficient Panoptic Segmentation code — Compare
PanNuke (4 rows) LKCell LKCell: Efficient Cell Nuclei Instance Segmentation with Large... code — Compare
ScanNet (4 rows) OneFormer3D OneFormer3D: One Transformer for Unified Point Cloud Segmentation code — Compare
PASTIS (3 rows) Exchanger+Mask2Former Revisiting the Encoding of Satellite Image Time Series code — Compare
COCO panoptic (2 rows) VAN-B6* Visual Attention Network code Syntology official (archive's flag): 1 ran · 6 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified Compare
MUSES: MUlti-SEnsor Semantic perception dataset (2 rows) CAFuser (Swin-T) CAFuser: Condition-Aware Multimodal Fusion for Robust Semantic... code Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified Compare
NYU Depth v2 (2 rows) EMSANet (2x ResNet-34 NBt1D, PanopticNDT version, finetuned) PanopticNDT: Efficient and Robust Panoptic Mapping code Syntology community repositories only · 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 7 unverified Compare
SemanticKITTI (2 rows) P3Former Position-Guided Point Cloud Panoptic Segmentation Transformer code — Compare
ADE20K (1 row) MasQCLIP MasQCLIP for Open-Vocabulary Universal Image Segmentation code — Compare
DALES (1 row) SuperCluster Scalable 3D Panoptic Segmentation As Superpoint Graph Clustering code — Compare
Hypersim (1 row) EMSANet (2x ResNet-34 NBt1D) PanopticNDT: Efficient and Robust Panoptic Mapping code Syntology community repositories only · 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 7 unverified Compare
KITTI-360 (1 row) SuperCluster Scalable 3D Panoptic Segmentation As Superpoint Graph Clustering code — Compare
Panoptic nuScenes val (1 row) PolarSeg-Panoptic Panoptic nuScenes: A Large-Scale Benchmark for LiDAR Panoptic... code — Compare
Panoptic nuScenes test (1 row) (AF)2-S3Net + CenterPoint Panoptic nuScenes: A Large-Scale Benchmark for LiDAR Panoptic... code — Compare
PASTIS-R (1 row) Early Fusion Multi-Modal Temporal Attention Models for Crop Mapping from... code — Compare
S3DIS (1 row) SuperCluster Scalable 3D Panoptic Segmentation As Superpoint Graph Clustering code — Compare
SUN-RGBD (1 row) EMSANet Efficient Multi-Task RGB-D Scene Analysis for Indoor Environments code — Compare

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

37 datasets whose archive record lists this task, ordered by the archive's paper count. 30 shown of 37 until expanded.

Subtasks archive 2025-07-28

2 subtasks in the archive's task tree.

Parent tasks archive 2025-07-28

Papers archive 2025-07-28

archive papers tagged: 462 · with a code link: 257 · where Syntology ran a sample: 88 (76 with a run with no instrument failure, 12 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (88 of 462 tagged: 76 with a run with no instrument failure, 12 where every run was a failure of Syntology's instrument)

Page 1 of 5: papers 1 to 100 of 462, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.

Papers without a page here are shown as plain text. A Syntology line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.

Syntology lines on 45 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

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