{"url":"/task/scene-understanding","name":"Scene Understanding","slug":"scene-understanding","description_markdown":"Scene understanding involves interpreting the visual information of a scene, including objects, their spatial relationships, and the overall layout. It goes beyond simple object recognition by considering the context and how objects relate to each other and the environment.","categories":[{"name":"Computer Vision","url":"/area/computer-vision"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":1723,"papers_with_code":720,"benchmarks":3,"benchmark_tables_in_archive":3,"benchmark_tables_shown":3,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":43,"subtasks":6,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/scene-understanding-on-semantic-scene","slug":"scene-understanding-on-semantic-scene","dataset":"Semantic Scene Understanding Challenge (passive actuation & ground-truth localisation)","dataset_url":null,"rows_in_archive":3,"metrics":["OMQ","avg_pairwise","avg_label","avg_spatial","avg_fp_quality"],"first_row_in_archive_order":{"model":"ACRV Baseline","paper_title":null,"paper_url":null,"paper_date":"","arxiv_id":null,"code_links":[],"syntology":null}},{"leaderboard":"/sota/scene-understanding-on-ade20k-val-1","slug":"scene-understanding-on-ade20k-val-1","dataset":"ADE20K val","dataset_url":"/dataset/ade20k","rows_in_archive":1,"metrics":["Mean IoU"],"first_row_in_archive_order":{"model":"CPN(ResNet-101)","paper_title":"Context Prior for Scene Segmentation","paper_url":"/paper/context-prior-for-scene-segmentation","paper_date":"2020-04-03","arxiv_id":"2004.01547","code_links":[{"title":"ycszen/ContextPrior","url":"https://github.com/ycszen/ContextPrior"},{"title":"AndPuQing/ContextPrior_Paddle","url":"https://github.com/AndPuQing/ContextPrior_Paddle"}],"syntology":null}},{"leaderboard":"/sota/scene-understanding-on-semantic-scene-1","slug":"scene-understanding-on-semantic-scene-1","dataset":"Semantic Scene Understanding Challenge (active actuation & ground-truth localisation)","dataset_url":null,"rows_in_archive":1,"metrics":["OMQ","avg_pairwise","avg_label","avg_spatial","avg_fp_quality"],"first_row_in_archive_order":{"model":"ACRV Baseline","paper_title":null,"paper_url":null,"paper_date":"","arxiv_id":null,"code_links":[],"syntology":null}}],"datasets":[{"url":"/dataset/ade20k","name":"ADE20K","full_name":"","num_papers_in_archive":1213},{"url":"/dataset/rellis-3d","name":"RELLIS-3D","full_name":null,"num_papers_in_archive":57},{"url":"/dataset/uavid","name":"UAVid","full_name":"","num_papers_in_archive":54},{"url":"/dataset/music-avqa","name":"MUSIC-AVQA","full_name":"","num_papers_in_archive":51},{"url":"/dataset/3r-scan","name":"3RScan","full_name":"","num_papers_in_archive":44},{"url":"/dataset/kitti-road","name":"KITTI Road","full_name":"","num_papers_in_archive":41},{"url":"/dataset/suim","name":"SUIM","full_name":"Segmentation of Underwater IMagery","num_papers_in_archive":34},{"url":"/dataset/sensaturban","name":"SensatUrban","full_name":"","num_papers_in_archive":28},{"url":"/dataset/radiate","name":"RADIATE","full_name":"RAdar Dataset In Adverse weaThEr","num_papers_in_archive":24},{"url":"/dataset/toronto-3d","name":"Toronto-3D","full_name":null,"num_papers_in_archive":24},{"url":"/dataset/shapestacks","name":"ShapeStacks","full_name":"","num_papers_in_archive":22},{"url":"/dataset/mlrsnet","name":"MLRSNet","full_name":"","num_papers_in_archive":17},{"url":"/dataset/scenenet","name":"SceneNet","full_name":"SceneNet","num_papers_in_archive":17},{"url":"/dataset/dada-2000","name":"DADA-2000","full_name":"","num_papers_in_archive":16},{"url":"/dataset/dublincity","name":"DublinCity","full_name":"","num_papers_in_archive":15},{"url":"/dataset/irs","name":"IRS","full_name":"Indoor Robotics Stereo","num_papers_in_archive":12},{"url":"/dataset/dawn","name":"DAWN","full_name":"","num_papers_in_archive":10},{"url":"/dataset/deepscores","name":"DeepScores","full_name":"","num_papers_in_archive":10},{"url":"/dataset/airs","name":"AIRS","full_name":"","num_papers_in_archive":9},{"url":"/dataset/pascal-panoptic-parts","name":"Pascal Panoptic Parts","full_name":"","num_papers_in_archive":9},{"url":"/dataset/psi-ava","name":"PSI-AVA","full_name":"","num_papers_in_archive":9},{"url":"/dataset/cityscapes-panoptic-parts","name":"Cityscapes Panoptic Parts","full_name":"","num_papers_in_archive":8},{"url":"/dataset/deeploc","name":"DeepLoc","full_name":"DeepLoc","num_papers_in_archive":8},{"url":"/dataset/synpick","name":"SynPick","full_name":"","num_papers_in_archive":7},{"url":"/dataset/a-large-dataset-of-object-scans","name":"A Large Dataset of Object Scans","full_name":"A Large Dataset of Object Scans","num_papers_in_archive":5},{"url":"/dataset/hhoi","name":"HHOI","full_name":"","num_papers_in_archive":5},{"url":"/dataset/sel","name":"SEL","full_name":"","num_papers_in_archive":5},{"url":"/dataset/www-crowd","name":"WWW Crowd","full_name":"WWW Crowd","num_papers_in_archive":5},{"url":"/dataset/aerorit","name":"AeroRIT","full_name":"","num_papers_in_archive":4},{"url":"/dataset/cds2k","name":"CDS2K","full_name":"","num_papers_in_archive":4},{"url":"/dataset/videonavqa","name":"VideoNavQA","full_name":"","num_papers_in_archive":4},{"url":"/dataset/mlm","name":"MLM","full_name":"","num_papers_in_archive":3},{"url":"/dataset/mod","name":"MOD++","full_name":"","num_papers_in_archive":3},{"url":"/dataset/raileye3d-dataset","name":"RailEye3D Dataset","full_name":"","num_papers_in_archive":3},{"url":"/dataset/instaorder","name":"InstaOrder","full_name":"","num_papers_in_archive":2},{"url":"/dataset/stanford-ecm","name":"Stanford-ECM","full_name":"Stanford-ECM","num_papers_in_archive":2},{"url":"/dataset/3threedworld","name":"3ThreeDWorld","full_name":"3ThreeDWorld","num_papers_in_archive":1},{"url":"/dataset/apron-dataset","name":"Apron Dataset","full_name":"","num_papers_in_archive":1},{"url":"/dataset/coqe","name":"COQE","full_name":"Containers Of liQuid contEnt","num_papers_in_archive":1},{"url":"/dataset/deeploccross","name":"DeepLocCross","full_name":"DeepLocCross","num_papers_in_archive":1},{"url":"/dataset/indoor-and-outdoor-dfd-dataset","name":"Indoor and outdoor DFD dataset","full_name":"","num_papers_in_archive":1},{"url":"/dataset/navigationnet","name":"NavigationNet","full_name":"","num_papers_in_archive":1},{"url":"/dataset/panoramic-video-panoptic-segmentation-dataset","name":"Panoramic Video Panoptic Segmentation Dataset","full_name":"","num_papers_in_archive":1}],"subtasks":[{"url":"/task/3d-room-layouts-from-a-single-rgb-panorama","name":"3D Room Layouts From A Single RGB Panorama"},{"url":"/task/light-source-estimation","name":"Outdoor Light Source Estimation"},{"url":"/task/lighting-estimation","name":"Lighting Estimation"},{"url":"/task/road-scene-understanding","name":"road scene understanding"},{"url":"/task/video-semantic-segmentation","name":"Video Semantic Segmentation"},{"url":"/task/visual-relationship-detection","name":"Visual Relationship Detection"}],"parent_tasks":[{"url":"/task/scene-parsing","name":"Scene Parsing"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":720,"tagged_in_all":1723,"items":[{"url":"/paper/segnet-a-deep-convolutional-encoder-decoder","title":"SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation","date":"2015-11-02","arxiv_id":"1511.00561","repositories_listed":74,"syntology":{"n":44,"n_ran":9,"n_unverified":35,"n_pointer_only":10}},{"url":"/paper/microsoft-coco-common-objects-in-context","title":"Microsoft COCO: Common Objects in Context","date":"2014-05-01","arxiv_id":"1405.0312","repositories_listed":38,"syntology":{"n":7,"n_ran":1,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/unified-perceptual-parsing-for-scene","title":"Unified Perceptual Parsing for Scene Understanding","date":"2018-07-26","arxiv_id":"1807.10221","repositories_listed":25,"syntology":{"n":28,"n_ran":9,"n_unverified":19,"n_pointer_only":0}},{"url":"/paper/bayesian-segnet-model-uncertainty-in-deep","title":"Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understanding","date":"2015-11-09","arxiv_id":"1511.02680","repositories_listed":21,"syntology":{"n":18,"n_ran":0,"n_unverified":18,"n_pointer_only":0}},{"url":"/paper/digging-into-self-supervised-monocular-depth","title":"Digging Into Self-Supervised Monocular Depth Estimation","date":"2018-06-04","arxiv_id":"1806.01260","repositories_listed":15,"syntology":{"n":24,"n_ran":17,"n_unverified":7,"n_pointer_only":6}},{"url":"/paper/linknet-exploiting-encoder-representations","title":"LinkNet: Exploiting Encoder Representations for Efficient Semantic Segmentation","date":"2017-06-14","arxiv_id":"1707.03718","repositories_listed":14,"syntology":{"n":4,"n_ran":3,"n_unverified":1,"n_pointer_only":2}},{"url":"/paper/erfnet-efficient-residual-factorized-convnet","title":"ERFNet: Efficient Residual Factorized ConvNet for Real-time Semantic Segmentation","date":"2017-10-09","arxiv_id":null,"repositories_listed":13,"syntology":null},{"url":"/paper/spatial-as-deep-spatial-cnn-for-traffic-scene","title":"Spatial As Deep: Spatial CNN for Traffic Scene Understanding","date":"2017-12-17","arxiv_id":"1712.06080","repositories_listed":9,"syntology":null},{"url":"/paper/decentralized-distributed-ppo-solving","title":"DD-PPO: Learning Near-Perfect PointGoal Navigators from 2.5 Billion Frames","date":"2019-11-01","arxiv_id":"1911.00357","repositories_listed":8,"syntology":null},{"url":"/paper/floodnet-a-high-resolution-aerial-imagery","title":"FloodNet: A High Resolution Aerial Imagery Dataset for Post Flood Scene Understanding","date":"2020-12-05","arxiv_id":"2012.02951","repositories_listed":7,"syntology":null},{"url":"/paper/resunet-a-a-deep-learning-framework-for","title":"ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data","date":"2019-04-01","arxiv_id":"1904.00592","repositories_listed":7,"syntology":{"n":13,"n_ran":1,"n_unverified":12,"n_pointer_only":0}},{"url":"/paper/multi-task-learning-as-multi-objective","title":"Multi-Task Learning as Multi-Objective Optimization","date":"2018-10-10","arxiv_id":"1810.04650","repositories_listed":7,"syntology":{"n":19,"n_ran":2,"n_unverified":17,"n_pointer_only":0}},{"url":"/paper/groupvit-semantic-segmentation-emerges-from","title":"GroupViT: Semantic Segmentation Emerges from Text Supervision","date":"2022-02-22","arxiv_id":"2202.11094","repositories_listed":6,"syntology":null},{"url":"/paper/part-a2-net-3d-part-aware-and-aggregation","title":"From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Network","date":"2019-07-08","arxiv_id":"1907.03670","repositories_listed":6,"syntology":{"n":8,"n_ran":1,"n_unverified":7,"n_pointer_only":0}},{"url":"/paper/multi-path-segmentation-network","title":"ShelfNet for Fast Semantic Segmentation","date":"2018-11-27","arxiv_id":"1811.11254","repositories_listed":6,"syntology":null},{"url":"/paper/boundary-seeking-generative-adversarial","title":"Boundary-Seeking Generative Adversarial Networks","date":"2017-02-27","arxiv_id":"1702.08431","repositories_listed":6,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/towards-ghost-free-shadow-removal-via-dual","title":"Towards Ghost-free Shadow Removal via Dual Hierarchical Aggregation Network and Shadow Matting GAN","date":"2019-11-20","arxiv_id":"1911.08718","repositories_listed":5,"syntology":null},{"url":"/paper/a-dataset-for-semantic-segmentation-of-point","title":"SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences","date":"2019-04-02","arxiv_id":"1904.01416","repositories_listed":5,"syntology":{"n":8,"n_ran":1,"n_unverified":7,"n_pointer_only":0}},{"url":"/paper/adapting-deep-network-features-to-capture","title":"Adapting Deep Network Features to Capture Psychological Representations","date":"2016-08-06","arxiv_id":"1608.02164","repositories_listed":5,"syntology":null},{"url":"/paper/repurposing-diffusion-based-image-generators","title":"Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation","date":"2023-12-04","arxiv_id":"2312.02145","repositories_listed":4,"syntology":{"n":26,"n_ran":14,"n_unverified":12,"n_pointer_only":0}},{"url":"/paper/panopticndt-efficient-and-robust-panoptic","title":"PanopticNDT: Efficient and Robust Panoptic Mapping","date":"2023-09-24","arxiv_id":"2309.13635","repositories_listed":4,"syntology":{"n":12,"n_ran":0,"n_unverified":12,"n_pointer_only":0}},{"url":"/paper/p2t-pyramid-pooling-transformer-for-scene","title":"P2T: Pyramid Pooling Transformer for Scene Understanding","date":"2021-06-22","arxiv_id":"2106.12011","repositories_listed":4,"syntology":null},{"url":"/paper/multi-view-radar-semantic-segmentation","title":"Multi-View Radar Semantic Segmentation","date":"2021-03-30","arxiv_id":"2103.16214","repositories_listed":4,"syntology":{"n":13,"n_ran":9,"n_unverified":4,"n_pointer_only":13}},{"url":"/paper/tracking-pedestrian-heads-in-dense-crowd","title":"Tracking Pedestrian Heads in Dense Crowd","date":"2021-03-24","arxiv_id":"2103.13516","repositories_listed":4,"syntology":null},{"url":"/paper/cityscapes-panoptic-parts-and-pascal-panoptic","title":"Cityscapes-Panoptic-Parts and PASCAL-Panoptic-Parts datasets for Scene Understanding","date":"2020-04-16","arxiv_id":"2004.07944","repositories_listed":4,"syntology":{"n":6,"n_ran":2,"n_unverified":4,"n_pointer_only":0}},{"url":"/paper/pointgroup-dual-set-point-grouping-for-3d","title":"PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation","date":"2020-04-03","arxiv_id":"2004.01658","repositories_listed":4,"syntology":{"n":5,"n_ran":0,"n_unverified":5,"n_pointer_only":0}},{"url":"/paper/visual-semantic-graph-attention-network-for","title":"Visual-Semantic Graph Attention Networks for Human-Object Interaction Detection","date":"2020-01-07","arxiv_id":"2001.02302","repositories_listed":4,"syntology":null},{"url":"/paper/language-based-colorization-of-scene-sketches","title":"Language-based Colorization of Scene Sketches","date":"2019-11-17","arxiv_id":null,"repositories_listed":4,"syntology":null},{"url":"/paper/m3d-rpn-monocular-3d-region-proposal-network","title":"M3D-RPN: Monocular 3D Region Proposal Network for Object Detection","date":"2019-07-13","arxiv_id":"1907.06038","repositories_listed":4,"syntology":{"n":21,"n_ran":1,"n_unverified":20,"n_pointer_only":0}},{"url":"/paper/multi-resolution-multi-modal-sensor-fusion","title":"Multi-Resolution Multi-Modal Sensor Fusion For Remote Sensing Data With Label Uncertainty","date":"2018-05-02","arxiv_id":"1805.00930","repositories_listed":4,"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}}],"syntology_records":18,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}