{"url":"/sota/3d-semantic-scene-completion-on-nyuv2","task":{"name":"3D Semantic Scene Completion","url":"/task/3d-semantic-scene-completion","note":null},"dataset":{"name":"NYUv2","url":"/dataset/nyuv2"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"This task was introduced in \"Semantic Scene Completion from a Single Depth Image\" (https://arxiv.org/abs/1611.08974) at CVPR 2017 . The target is to infer the dense 3D voxelized semantic scene from an incompleted 3D input (e.g. point cloud, depth map) and an optional RGB image.  A recent summary can be found in the paper  \"3D Semantic Scene Completion: a Survey\" (https://arxiv.org/abs/2103.07466), published at IJCV 2021.","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["mIoU"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"mIoU":null}},"counts":{"rows":28,"rows_with_code":13,"rows_with_paper_page":28,"rows_dated":28,"rows_using_additional_data":5},"rows":[{"rank_in_archive_order":1,"model":"SG-SSC","metrics":{"mIoU":"55.4"},"uses_additional_data":false,"paper_date":"2024-09-12","paper":"/paper/2d-semantic-guided-semantic-scene-completion","paper_url":"https://doi.org/10.1007/s11263-024-02244-y","paper_title":"2D Semantic-Guided Semantic Scene Completion","code":"https://github.com/aipixel/SG-SSC","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"SISNet","metrics":{"mIoU":"52.4"},"uses_additional_data":false,"paper_date":"2021-04-08","paper":"/paper/semantic-scene-completion-via-integrating","paper_url":"https://arxiv.org/abs/2104.03640v2","paper_title":"Semantic Scene Completion via Integrating Instances and Scene in-the-Loop","code":"https://github.com/yjcaimeow/SISNet","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":3,"model":"SPAwN (SUNCG pretraining)","metrics":{"mIoU":"49.9"},"uses_additional_data":false,"paper_date":"2021-11-26","paper":"/paper/data-augmented-3d-semantic-scene-completion","paper_url":"https://arxiv.org/abs/2111.13309v1","paper_title":"Data Augmented 3D Semantic Scene Completion with 2D Segmentation Priors","code":"https://gitlab.com/UnBVision/spawn","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"SPAwN","metrics":{"mIoU":"48"},"uses_additional_data":false,"paper_date":"2021-11-26","paper":"/paper/data-augmented-3d-semantic-scene-completion","paper_url":"https://arxiv.org/abs/2111.13309v1","paper_title":"Data Augmented 3D Semantic Scene Completion with 2D Segmentation Priors","code":"https://gitlab.com/UnBVision/spawn","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"Point-Voxel Aggregation Network","metrics":{"mIoU":"46"},"uses_additional_data":false,"paper_date":"2021-12-24","paper":"/paper/not-all-voxels-are-equal-semantic-scene","paper_url":"https://arxiv.org/abs/2112.12925v2","paper_title":"Not All Voxels Are Equal: Semantic Scene Completion from the Point-Voxel Perspective","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"CCPNet (SUNCG pretraining)","metrics":{"mIoU":"41.3"},"uses_additional_data":true,"paper_date":"2019-08-01","paper":"/paper/cascaded-context-pyramid-for-full-resolution","paper_url":"https://arxiv.org/abs/1908.00382v1","paper_title":"Cascaded Context Pyramid for Full-Resolution 3D Semantic Scene Completion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"3DSketch","metrics":{"mIoU":"41.1"},"uses_additional_data":false,"paper_date":"2020-03-31","paper":"/paper/3d-sketch-aware-semantic-scene-completion-via","paper_url":"https://arxiv.org/abs/2003.14052v1","paper_title":"3D Sketch-aware Semantic Scene Completion via Semi-supervised Structure Prior","code":"https://github.com/charlesCXK/3D-SketchAware-SSC","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":2,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"ForkNet (SUNCG pretraining)","metrics":{"mIoU":"37.1"},"uses_additional_data":true,"paper_date":"2019-09-03","paper":"/paper/forknet-multi-branch-volumetric-semantic","paper_url":"https://arxiv.org/abs/1909.01106v1","paper_title":"ForkNet: Multi-branch Volumetric Semantic Completion from a Single Depth Image","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"IPF-SPCNet: Semantic point completion network for 3D\nsemantic scene completion","metrics":{"mIoU":"35.1"},"uses_additional_data":false,"paper_date":"2020-08-29","paper":"/paper/semantic-point-completion-network-for-3d","paper_url":"https://ecai2020.eu/papers/123_paper.pdf","paper_title":"Semantic Point Completion Network for 3D Semantic Scene Completion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":10,"model":"Real-time semantic scene completion via feature aggregation and conditioned prediction","metrics":{"mIoU":"34.4"},"uses_additional_data":false,"paper_date":"2021-03-12","paper":"/paper/3d-semantic-scene-completion-a-survey","paper_url":"https://arxiv.org/abs/2103.07466v3","paper_title":"3D Semantic Scene Completion: a Survey","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":11,"model":"TS3D","metrics":{"mIoU":"34.1"},"uses_additional_data":false,"paper_date":"2018-04-10","paper":"/paper/two-stream-3d-semantic-scene-completion","paper_url":"https://arxiv.org/abs/1804.03550v4","paper_title":"Two Stream 3D Semantic Scene Completion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"EdgeNet (SUNCG pretraining)","metrics":{"mIoU":"33.7"},"uses_additional_data":true,"paper_date":"2021-03-12","paper":"/paper/3d-semantic-scene-completion-a-survey","paper_url":"https://arxiv.org/abs/2103.07466v3","paper_title":"3D Semantic Scene Completion: a Survey","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"AIC-Net","metrics":{"mIoU":"33.3"},"uses_additional_data":false,"paper_date":"2020-04-05","paper":"/paper/anisotropic-convolutional-networks-for-3d","paper_url":"https://arxiv.org/abs/2004.02122v1","paper_title":"Anisotropic Convolutional Networks for 3D Semantic Scene Completion","code":"https://github.com/waterljwant/SSC","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"AMFNet","metrics":{"mIoU":"33"},"uses_additional_data":false,"paper_date":"2020-03-31","paper":"/paper/attention-based-multi-modal-fusion-network","paper_url":"https://arxiv.org/abs/2003.13910v2","paper_title":"Attention-based Multi-modal Fusion Network for Semantic Scene Completion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":15,"model":"GRFNet","metrics":{"mIoU":"32.9"},"uses_additional_data":false,"paper_date":"2020-02-17","paper":"/paper/3d-gated-recurrent-fusion-for-semantic-scene","paper_url":"https://arxiv.org/abs/2002.07269v1","paper_title":"3D Gated Recurrent Fusion for Semantic Scene Completion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":16,"model":"VD-CRF: Semantic scene completion with dense\nCRF from a single depth image. (SUNCG pretraining)","metrics":{"mIoU":"31.8"},"uses_additional_data":true,"paper_date":"2021-03-12","paper":"/paper/3d-semantic-scene-completion-a-survey","paper_url":"https://arxiv.org/abs/2103.07466v3","paper_title":"3D Semantic Scene Completion: a Survey","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":17,"model":"Am2fnet: Attention-based multiscale\n& multi-modality fused network","metrics":{"mIoU":"31.7"},"uses_additional_data":false,"paper_date":"2021-03-12","paper":"/paper/3d-semantic-scene-completion-a-survey","paper_url":"https://arxiv.org/abs/2103.07466v3","paper_title":"3D Semantic Scene Completion: a Survey","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":18,"model":"OccDepth","metrics":{"mIoU":"30.9"},"uses_additional_data":false,"paper_date":"2023-02-27","paper":"/paper/occdepth-a-depth-aware-method-for-3d-semantic","paper_url":"https://arxiv.org/abs/2302.13540v1","paper_title":"OccDepth: A Depth-Aware Method for 3D Semantic Scene Completion","code":"https://github.com/megvii-research/occdepth","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":3,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"SSCNet (SUNCG pretraining)","metrics":{"mIoU":"30.5"},"uses_additional_data":true,"paper_date":"2016-11-28","paper":"/paper/semantic-scene-completion-from-a-single-depth","paper_url":"http://arxiv.org/abs/1611.08974v1","paper_title":"Semantic Scene Completion from a Single Depth Image","code":"https://github.com/facebookresearch/House3D","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":20,"model":"DDRNet","metrics":{"mIoU":"30.4"},"uses_additional_data":false,"paper_date":"2019-03-02","paper":"/paper/rgbd-based-dimensional-decomposition-residual","paper_url":"http://arxiv.org/abs/1903.00620v2","paper_title":"RGBD Based Dimensional Decomposition Residual Network for 3D Semantic Scene Completion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":21,"model":"VVNet","metrics":{"mIoU":"29.3"},"uses_additional_data":false,"paper_date":"2018-06-14","paper":"/paper/view-volume-network-for-semantic-scene","paper_url":"http://arxiv.org/abs/1806.05361v1","paper_title":"View-volume Network for Semantic Scene Completion from a Single Depth Image","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":22,"model":"LMSCNet-SS","metrics":{"mIoU":"28.4"},"uses_additional_data":false,"paper_date":"2020-08-24","paper":"/paper/lmscnet-lightweight-multiscale-3d-semantic","paper_url":"https://arxiv.org/abs/2008.10559v2","paper_title":"LMSCNet: Lightweight Multiscale 3D Semantic Completion","code":"https://github.com/astra-vision/lmscnet","n_code_links":3,"syntology":null},{"rank_in_archive_order":23,"model":"EdgeNet","metrics":{"mIoU":"27.8"},"uses_additional_data":false,"paper_date":"2019-08-08","paper":"/paper/edgenet-semantic-scene-completion-from-rgb-d","paper_url":"https://arxiv.org/abs/1908.02893v2","paper_title":"EdgeNet: Semantic Scene Completion from a Single RGB-D Image","code":"https://gitlab.com/UnBVision/edgenet","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"Guedes et al.","metrics":{"mIoU":"27.5"},"uses_additional_data":false,"paper_date":"2018-02-13","paper":"/paper/semantic-scene-completion-combining-colour","paper_url":"http://arxiv.org/abs/1802.04735v1","paper_title":"Semantic Scene Completion Combining Colour and Depth: preliminary experiments","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":25,"model":"MonoScene (RGB input only)","metrics":{"mIoU":"26.94"},"uses_additional_data":false,"paper_date":"2021-12-01","paper":"/paper/monoscene-monocular-3d-semantic-scene","paper_url":"https://arxiv.org/abs/2112.00726v2","paper_title":"MonoScene: Monocular 3D Semantic Scene Completion","code":"https://github.com/astra-vision/monoscene","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":3,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"ESSCNet","metrics":{"mIoU":"26.7"},"uses_additional_data":false,"paper_date":"2019-07-11","paper":"/paper/efficient-semantic-scene-completion-network-1","paper_url":"https://arxiv.org/abs/1907.05091v1","paper_title":"Efficient Semantic Scene Completion Network with Spatial Group Convolution","code":"https://github.com/zjhthu/SGC-Release","n_code_links":1,"syntology":null},{"rank_in_archive_order":27,"model":"SSCNet","metrics":{"mIoU":"24.7"},"uses_additional_data":false,"paper_date":"2016-11-28","paper":"/paper/semantic-scene-completion-from-a-single-depth","paper_url":"http://arxiv.org/abs/1611.08974v1","paper_title":"Semantic Scene Completion from a Single Depth Image","code":"https://github.com/facebookresearch/House3D","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":28,"model":"3D semantic scene completion\nfrom a single depth image using adversarial training","metrics":{"mIoU":"22.7"},"uses_additional_data":false,"paper_date":"2021-03-12","paper":"/paper/3d-semantic-scene-completion-a-survey","paper_url":"https://arxiv.org/abs/2103.07466v3","paper_title":"3D Semantic Scene Completion: a Survey","code":null,"n_code_links":0,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,795 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6795,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2785},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":6,"rows_with_any_sample_ran":6,"distinct_papers_with_graph_line":5,"distinct_papers_with_any_sample_ran":5,"samples_over_distinct_papers":{"n_ran":11,"n_unverified":8,"n_samples":19,"n_pointer_only_licence":4,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":12,"n_unverified":8,"n_samples":20,"n_pointer_only_licence":5,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}