{"url":"/task/real-time-semantic-segmentation","name":"Real-Time Semantic Segmentation","slug":"real-time-semantic-segmentation","description_markdown":"Semantic Segmentation is a computer vision task that involves assigning a semantic label to each pixel in an image. In **Real-Time Semantic Segmentation**, the goal is to perform this labeling quickly and accurately in real-time, allowing for the segmentation results to be used for tasks such as object recognition, scene understanding, and autonomous navigation.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [TorchSeg](https://github.com/ycszen/TorchSeg) )</span>","categories":[{"name":"Computer Code","url":"/area/computer-code"},{"name":"Computer Vision","url":"/area/computer-vision"},{"name":"Medical","url":"/area/medical"},{"name":"Robots","url":"/area/robots"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":145,"papers_with_code":98,"benchmarks":8,"benchmark_tables_in_archive":8,"benchmark_tables_shown":8,"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":12,"subtasks":0,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/real-time-semantic-segmentation-on-cityscapes","slug":"real-time-semantic-segmentation-on-cityscapes","dataset":"Cityscapes test","dataset_url":"/dataset/cityscapes","rows_in_archive":39,"metrics":["mIoU","Frame (fps)","Time (ms)"],"first_row_in_archive_order":{"model":"PIDNet-L","paper_title":"PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers","paper_url":"/paper/pidnet-a-real-time-semantic-segmentation","paper_date":"2022-06-04","arxiv_id":"2206.02066","code_links":[{"title":"XuJiacong/PIDNet","url":"https://github.com/XuJiacong/PIDNet"},{"title":"Darth-Kronos/PIDNet_TensorRT","url":"https://github.com/Darth-Kronos/PIDNet_TensorRT"},{"title":"hamidriasat/PIDNet","url":"https://github.com/hamidriasat/PIDNet"},{"title":"HengWeiBin/Oil-Polution-Dataset-with-PIDNet","url":"https://github.com/HengWeiBin/Oil-Polution-Dataset-with-PIDNet"},{"title":"Mahmood-Hussain/PIDNetTensorflow","url":"https://github.com/Mahmood-Hussain/PIDNetTensorflow"},{"title":"enot-autodl/lpcv-2023","url":"https://github.com/enot-autodl/lpcv-2023"}],"syntology":{"n":19,"n_ran":1,"n_unverified":18,"n_pointer_only":1}}},{"leaderboard":"/sota/real-time-semantic-segmentation-on-camvid","slug":"real-time-semantic-segmentation-on-camvid","dataset":"CamVid","dataset_url":"/dataset/camvid","rows_in_archive":29,"metrics":["mIoU","Frame (fps)","Time (ms)"],"first_row_in_archive_order":{"model":"PIDNet-M (Cityscapes-Pretrained)","paper_title":"PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers","paper_url":"/paper/pidnet-a-real-time-semantic-segmentation","paper_date":"2022-06-04","arxiv_id":"2206.02066","code_links":[{"title":"XuJiacong/PIDNet","url":"https://github.com/XuJiacong/PIDNet"},{"title":"Darth-Kronos/PIDNet_TensorRT","url":"https://github.com/Darth-Kronos/PIDNet_TensorRT"},{"title":"hamidriasat/PIDNet","url":"https://github.com/hamidriasat/PIDNet"},{"title":"HengWeiBin/Oil-Polution-Dataset-with-PIDNet","url":"https://github.com/HengWeiBin/Oil-Polution-Dataset-with-PIDNet"},{"title":"Mahmood-Hussain/PIDNetTensorflow","url":"https://github.com/Mahmood-Hussain/PIDNetTensorflow"},{"title":"enot-autodl/lpcv-2023","url":"https://github.com/enot-autodl/lpcv-2023"}],"syntology":{"n":19,"n_ran":1,"n_unverified":18,"n_pointer_only":1}}},{"leaderboard":"/sota/real-time-semantic-segmentation-on-cityscapes-1","slug":"real-time-semantic-segmentation-on-cityscapes-1","dataset":"Cityscapes val","dataset_url":"/dataset/cityscapes","rows_in_archive":24,"metrics":["mIoU","Frame (fps)","Time (ms)"],"first_row_in_archive_order":{"model":"PIDNet-L","paper_title":"PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers","paper_url":"/paper/pidnet-a-real-time-semantic-segmentation","paper_date":"2022-06-04","arxiv_id":"2206.02066","code_links":[{"title":"XuJiacong/PIDNet","url":"https://github.com/XuJiacong/PIDNet"},{"title":"Darth-Kronos/PIDNet_TensorRT","url":"https://github.com/Darth-Kronos/PIDNet_TensorRT"},{"title":"hamidriasat/PIDNet","url":"https://github.com/hamidriasat/PIDNet"},{"title":"HengWeiBin/Oil-Polution-Dataset-with-PIDNet","url":"https://github.com/HengWeiBin/Oil-Polution-Dataset-with-PIDNet"},{"title":"Mahmood-Hussain/PIDNetTensorflow","url":"https://github.com/Mahmood-Hussain/PIDNetTensorflow"},{"title":"enot-autodl/lpcv-2023","url":"https://github.com/enot-autodl/lpcv-2023"}],"syntology":{"n":19,"n_ran":1,"n_unverified":18,"n_pointer_only":1}}},{"leaderboard":"/sota/real-time-semantic-segmentation-on-nyu-depth-1","slug":"real-time-semantic-segmentation-on-nyu-depth-1","dataset":"NYU Depth v2","dataset_url":"/dataset/nyuv2","rows_in_archive":10,"metrics":["mIoU","Speed(ms/f)","Speed  (FPS)"],"first_row_in_archive_order":{"model":"AsymFormer","paper_title":"AsymFormer: Asymmetrical Cross-Modal Representation Learning for Mobile Platform Real-Time RGB-D Semantic Segmentation","paper_url":"/paper/asymformer-asymmetrical-cross-modal","paper_date":"2023-09-25","arxiv_id":"2309.14065","code_links":[{"title":"Fourier7754/AsymFormer","url":"https://github.com/Fourier7754/AsymFormer"}],"syntology":null}},{"leaderboard":"/sota/real-time-semantic-segmentation-on-cityscapes-3","slug":"real-time-semantic-segmentation-on-cityscapes-3","dataset":"Cityscapes","dataset_url":"/dataset/cityscapes","rows_in_archive":2,"metrics":["mIoU"],"first_row_in_archive_order":{"model":"S^2-FPN34","paper_title":"S$^2$-FPN: Scale-ware Strip Attention Guided Feature Pyramid Network for Real-time Semantic Segmentation","paper_url":"/paper/s-textsuperscript-2-fpn-scale-ware-strip","paper_date":"2022-06-15","arxiv_id":"2206.07298","code_links":[{"title":"mohamedac29/s2-fpn","url":"https://github.com/mohamedac29/s2-fpn"}],"syntology":null}},{"leaderboard":"/sota/real-time-semantic-segmentation-on-coco-stuff-1","slug":"real-time-semantic-segmentation-on-coco-stuff-1","dataset":"COCO-Stuff","dataset_url":"/dataset/coco-stuff","rows_in_archive":2,"metrics":["Frame (fps)","mIoU"],"first_row_in_archive_order":{"model":"BiSeNet V2-Large","paper_title":"BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation","paper_url":"/paper/bisenet-v2-bilateral-network-with-guided","paper_date":"2020-04-05","arxiv_id":"2004.02147","code_links":[{"title":"PaddlePaddle/PaddleSeg","url":"https://github.com/PaddlePaddle/PaddleSeg"},{"title":"CoinCheung/BiSeNet","url":"https://github.com/CoinCheung/BiSeNet"},{"title":"MaybeShewill-CV/bisenetv2-tensorflow","url":"https://github.com/MaybeShewill-CV/bisenetv2-tensorflow"},{"title":"zh320/realtime-semantic-segmentation-pytorch","url":"https://github.com/zh320/realtime-semantic-segmentation-pytorch"},{"title":"ycszen/BiSeNet","url":"https://github.com/ycszen/BiSeNet"},{"title":"hamidriasat/BiSeNetV2","url":"https://github.com/hamidriasat/BiSeNetV2"},{"title":"gymoon10/Instance-Segmentation-with-SpatialEmbedding-CA","url":"https://github.com/gymoon10/Instance-Segmentation-with-SpatialEmbedding-CA"}],"syntology":{"n":10,"n_ran":6,"n_unverified":4,"n_pointer_only":5}}},{"leaderboard":"/sota/real-time-semantic-segmentation-on-flame","slug":"real-time-semantic-segmentation-on-flame","dataset":"FLAME","dataset_url":"/dataset/flame","rows_in_archive":1,"metrics":["FPS","Mean Intersection over Union","Mean Pixel Accuracy"],"first_row_in_archive_order":{"model":"Fast DeepLabV3+","paper_title":"A Real-time Fire Segmentation Method Based on A Deep Learning Approach","paper_url":"/paper/a-real-time-fire-segmentation-method-based-on","paper_date":"2022-07-22","arxiv_id":null,"code_links":[{"title":"maidacundo/real-time-fire-segmentation-deep-learning","url":"https://github.com/maidacundo/real-time-fire-segmentation-deep-learning"}],"syntology":null}},{"leaderboard":"/sota/real-time-semantic-segmentation-on-helixnet","slug":"real-time-semantic-segmentation-on-helixnet","dataset":"HelixNet","dataset_url":"/dataset/helixnet","rows_in_archive":1,"metrics":["Inference Time (ms) (1/5 rotation)","mIoU (1/5 rotation)"],"first_row_in_archive_order":{"model":"Helix4D","paper_title":"Online Segmentation of LiDAR Sequences: Dataset and Algorithm","paper_url":"/paper/online-segmentation-of-lidar-sequences","paper_date":"2022-06-16","arxiv_id":"2206.08194","code_links":[{"title":"romainloiseau/Helix4D","url":"https://github.com/romainloiseau/Helix4D"}],"syntology":null}}],"datasets":[{"url":"/dataset/cityscapes","name":"Cityscapes","full_name":"","num_papers_in_archive":3702},{"url":"/dataset/nyuv2","name":"NYUv2","full_name":"NYU-Depth V2","num_papers_in_archive":986},{"url":"/dataset/coco-stuff","name":"COCO-Stuff","full_name":"Common Objects in COntext-stuff","num_papers_in_archive":338},{"url":"/dataset/camvid","name":"CamVid","full_name":"Cambridge-driving Labeled Video Database","num_papers_in_archive":227},{"url":"/dataset/kvasir-seg","name":"Kvasir-SEG","full_name":"","num_papers_in_archive":201},{"url":"/dataset/kvasir","name":"Kvasir","full_name":"The Kvasir Dataset","num_papers_in_archive":117},{"url":"/dataset/kvasir-instrument","name":"Kvasir-Instrument","full_name":"","num_papers_in_archive":19},{"url":"/dataset/flame","name":"FLAME","full_name":"Fire Luminosity Airborne-based Machine learning Evaluation","num_papers_in_archive":12},{"url":"/dataset/helixnet","name":"HelixNet","full_name":"HelixNet: A Dataset for Online LiDAR Segmentation","num_papers_in_archive":3},{"url":"/dataset/kvasircapsule-seg","name":"KvasirCapsule-SEG","full_name":"","num_papers_in_archive":3},{"url":"/dataset/medico-automatic-polyp-segmentation-challenge","name":"Medico automatic polyp segmentation challenge (dataset)","full_name":"","num_papers_in_archive":3},{"url":"/dataset/endotect-polyp-segmentation","name":"Endotect Polyp Segmentation Challenge Dataset","full_name":"","num_papers_in_archive":2}],"subtasks":[],"parent_tasks":[{"url":"/task/semantic-segmentation","name":"Semantic Segmentation"}],"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":98,"tagged_in_all":145,"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/pyramid-scene-parsing-network","title":"Pyramid Scene Parsing Network","date":"2016-12-04","arxiv_id":"1612.01105","repositories_listed":67,"syntology":{"n":29,"n_ran":7,"n_unverified":22,"n_pointer_only":5}},{"url":"/paper/enet-a-deep-neural-network-architecture-for","title":"ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation","date":"2016-06-07","arxiv_id":"1606.02147","repositories_listed":49,"syntology":{"n":30,"n_ran":5,"n_unverified":25,"n_pointer_only":0}},{"url":"/paper/fully-convolutional-networks-for-semantic","title":"Fully Convolutional Networks for Semantic Segmentation","date":"2016-05-20","arxiv_id":"1605.06211","repositories_listed":37,"syntology":null},{"url":"/paper/hardnet-a-low-memory-traffic-network","title":"HarDNet: A Low Memory Traffic Network","date":"2019-09-03","arxiv_id":"1909.00948","repositories_listed":24,"syntology":null},{"url":"/paper/fast-scnn-fast-semantic-segmentation-network","title":"Fast-SCNN: Fast Semantic Segmentation Network","date":"2019-02-12","arxiv_id":"1902.04502","repositories_listed":24,"syntology":{"n":32,"n_ran":0,"n_unverified":32,"n_pointer_only":0}},{"url":"/paper/bisenet-bilateral-segmentation-network-for","title":"BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation","date":"2018-08-02","arxiv_id":"1808.00897","repositories_listed":21,"syntology":{"n":18,"n_ran":6,"n_unverified":12,"n_pointer_only":1}},{"url":"/paper/solov2-dynamic-faster-and-stronger","title":"SOLOv2: Dynamic and Fast Instance Segmentation","date":"2020-03-23","arxiv_id":"2003.10152","repositories_listed":18,"syntology":{"n":38,"n_ran":15,"n_unverified":23,"n_pointer_only":24}},{"url":"/paper/icnet-for-real-time-semantic-segmentation-on","title":"ICNet for Real-Time Semantic Segmentation on High-Resolution Images","date":"2017-04-27","arxiv_id":"1704.08545","repositories_listed":18,"syntology":null},{"url":"/paper/lite-hrnet-a-lightweight-high-resolution","title":"Lite-HRNet: A Lightweight High-Resolution Network","date":"2021-04-13","arxiv_id":"2104.06403","repositories_listed":17,"syntology":{"n":29,"n_ran":12,"n_unverified":17,"n_pointer_only":13}},{"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/espnetv2-a-light-weight-power-efficient-and","title":"ESPNetv2: A Light-weight, Power Efficient, and General Purpose Convolutional Neural Network","date":"2018-11-28","arxiv_id":"1811.11431","repositories_listed":10,"syntology":{"n":4,"n_ran":2,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/multi-scale-context-aggregation-by-dilated","title":"Multi-Scale Context Aggregation by Dilated Convolutions","date":"2015-11-23","arxiv_id":"1511.07122","repositories_listed":9,"syntology":{"n":8,"n_ran":2,"n_unverified":6,"n_pointer_only":2}},{"url":"/paper/deep-dual-resolution-networks-for-real-time","title":"Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes","date":"2021-01-15","arxiv_id":"2101.06085","repositories_listed":8,"syntology":null},{"url":"/paper/espnet-efficient-spatial-pyramid-of-dilated","title":"ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation","date":"2018-03-19","arxiv_id":"1803.06815","repositories_listed":8,"syntology":{"n":10,"n_ran":4,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/bisenet-v2-bilateral-network-with-guided","title":"BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation","date":"2020-04-05","arxiv_id":"2004.02147","repositories_listed":7,"syntology":{"n":10,"n_ran":6,"n_unverified":4,"n_pointer_only":5}},{"url":"/paper/pidnet-a-real-time-semantic-segmentation","title":"PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers","date":"2022-06-04","arxiv_id":"2206.02066","repositories_listed":6,"syntology":{"n":19,"n_ran":1,"n_unverified":18,"n_pointer_only":1}},{"url":"/paper/rethinking-bisenet-for-real-time-semantic","title":"Rethinking BiSeNet For Real-time Semantic Segmentation","date":"2021-04-27","arxiv_id":"2104.13188","repositories_listed":6,"syntology":{"n":16,"n_ran":8,"n_unverified":8,"n_pointer_only":0}},{"url":"/paper/semantic-flow-for-fast-and-accurate-scene","title":"Semantic Flow for Fast and Accurate Scene Parsing","date":"2020-02-24","arxiv_id":"2002.10120","repositories_listed":6,"syntology":{"n":8,"n_ran":1,"n_unverified":7,"n_pointer_only":1}},{"url":"/paper/lednet-a-lightweight-encoder-decoder-network","title":"LEDNet: A Lightweight Encoder-Decoder Network for Real-Time Semantic Segmentation","date":"2019-05-07","arxiv_id":"1905.02423","repositories_listed":6,"syntology":null},{"url":"/paper/in-defense-of-pre-trained-imagenet","title":"In Defense of Pre-trained ImageNet Architectures for Real-time Semantic Segmentation of Road-driving Images","date":"2019-03-20","arxiv_id":"1903.08469","repositories_listed":6,"syntology":null},{"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/conditional-random-fields-as-recurrent-neural-1","title":"Conditional Random Fields as Recurrent Neural Networks","date":"2015-02-11","arxiv_id":"1502.03240","repositories_listed":6,"syntology":{"n":8,"n_ran":1,"n_unverified":7,"n_pointer_only":2}},{"url":"/paper/segnext-rethinking-convolutional-attention","title":"SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation","date":"2022-09-18","arxiv_id":"2209.08575","repositories_listed":5,"syntology":null},{"url":"/paper/efficient-dense-modules-of-asymmetric","title":"Efficient Dense Modules of Asymmetric Convolution for Real-Time Semantic Segmentation","date":"2018-09-17","arxiv_id":"1809.06323","repositories_listed":4,"syntology":null},{"url":"/paper/real-time-joint-semantic-segmentation-and","title":"Real-Time Joint Semantic Segmentation and Depth Estimation Using Asymmetric Annotations","date":"2018-09-13","arxiv_id":"1809.04766","repositories_listed":4,"syntology":{"n":6,"n_ran":6,"n_unverified":0,"n_pointer_only":6}},{"url":"/paper/full-resolution-residual-networks-for","title":"Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes","date":"2016-11-24","arxiv_id":"1611.08323","repositories_listed":4,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":3}},{"url":"/paper/pp-liteseg-a-superior-real-time-semantic","title":"PP-LiteSeg: A Superior Real-Time Semantic Segmentation Model","date":"2022-04-06","arxiv_id":"2204.02681","repositories_listed":3,"syntology":null},{"url":"/paper/rethink-dilated-convolution-for-real-time","title":"Rethinking Dilated Convolution for Real-time Semantic Segmentation","date":"2021-11-18","arxiv_id":"2111.09957","repositories_listed":3,"syntology":null},{"url":"/paper/nanonet-real-time-polyp-segmentation-in-video","title":"NanoNet: Real-Time Polyp Segmentation in Video Capsule Endoscopy and Colonoscopy","date":"2021-04-22","arxiv_id":"2104.11138","repositories_listed":3,"syntology":null}],"syntology_records":17,"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"}}