{"url":"/sota/3d-object-detection-on-nuscenes-camera-only","task":{"name":"3D Object Detection","url":"/task/3d-object-detection","note":null},"dataset":{"name":"nuScenes Camera Only","url":"/dataset/nuscenes"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"**3D Object Detection** is a task in computer vision where the goal is to identify and locate objects in a 3D environment based on their shape, location, and orientation. It involves detecting the presence of objects and determining their location in the 3D space in real-time. This task is crucial for applications such as autonomous vehicles, robotics, and augmented reality.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [AVOD](https://github.com/kujason/avod) )</span>","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":["NDS","Future Frame"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"NDS":null,"Future Frame":null}},"counts":{"rows":19,"rows_with_code":19,"rows_with_paper_page":19,"rows_dated":19,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"Far3D","metrics":{"Future Frame":"false","NDS":"68.7"},"uses_additional_data":false,"paper_date":"2023-08-18","paper":"/paper/far3d-expanding-the-horizon-for-surround-view","paper_url":"https://arxiv.org/abs/2308.09616v2","paper_title":"Far3D: Expanding the Horizon for Surround-view 3D Object Detection","code":"https://github.com/megvii-research/far3d","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":0,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":2,"model":"RayDN","metrics":{"Future Frame":"false","NDS":"68.6"},"uses_additional_data":false,"paper_date":"2024-02-06","paper":"/paper/beam-beta-distribution-ray-denoising-for","paper_url":"https://arxiv.org/abs/2402.03634v2","paper_title":"Ray Denoising: Depth-aware Hard Negative Sampling for Multi-view 3D Object Detection","code":"https://github.com/liewfeng/beam","n_code_links":2,"syntology":{"n_ran":9,"n_unverified":1,"n_samples":10,"n_pointer_only_licence":10}},{"rank_in_archive_order":3,"model":"HoP","metrics":{"Future Frame":"yes","NDS":"68.5"},"uses_additional_data":false,"paper_date":"2023-04-03","paper":"/paper/temporal-enhanced-training-of-multi-view-3d","paper_url":"https://arxiv.org/abs/2304.00967v1","paper_title":"Temporal Enhanced Training of Multi-view 3D Object Detector via Historical Object Prediction","code":"https://github.com/sense-x/hop","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":1,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"StreamPETR-Large","metrics":{"Future Frame":"false","NDS":"67.6"},"uses_additional_data":false,"paper_date":"2023-03-21","paper":"/paper/exploring-object-centric-temporal-modeling","paper_url":"https://arxiv.org/abs/2303.11926v2","paper_title":"Exploring Object-Centric Temporal Modeling for Efficient Multi-View 3D Object Detection","code":"https://github.com/exiawsh/streampetr","n_code_links":2,"syntology":null},{"rank_in_archive_order":5,"model":"SparseBEV (V2-99)","metrics":{"Future Frame":"yes","NDS":"67.5"},"uses_additional_data":false,"paper_date":"2023-08-18","paper":"/paper/sparsebev-high-performance-sparse-3d-object","paper_url":"https://arxiv.org/abs/2308.09244v2","paper_title":"SparseBEV: High-Performance Sparse 3D Object Detection from Multi-Camera Videos","code":"https://github.com/mcg-nju/sparsebev","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"VCD-A","metrics":{"Future Frame":"false","NDS":"67.2"},"uses_additional_data":false,"paper_date":"2023-10-24","paper":"/paper/leveraging-vision-centric-multi-modal-1","paper_url":"https://arxiv.org/abs/2310.15670v1","paper_title":"Leveraging Vision-Centric Multi-Modal Expertise for 3D Object Detection","code":"https://github.com/opendrivelab/birds-eye-view-perception","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":1,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"GeoBEV (V2-99)","metrics":{"NDS":"66.2"},"uses_additional_data":false,"paper_date":"2024-09-03","paper":"/paper/geobev-learning-geometric-bev-representation","paper_url":"https://arxiv.org/abs/2409.01816v2","paper_title":"GeoBEV: Learning Geometric BEV Representation for Multi-view 3D Object Detection","code":"https://github.com/mengtan00/geobev","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"BEVFormer v2 (InternImage-XL)","metrics":{"Future Frame":"yes","NDS":"63.4"},"uses_additional_data":false,"paper_date":"2022-11-18","paper":"/paper/bevformer-v2-adapting-modern-image-backbones","paper_url":"https://arxiv.org/abs/2211.10439v1","paper_title":"BEVFormer v2: Adapting Modern Image Backbones to Bird's-Eye-View Recognition via Perspective Supervision","code":"https://github.com/fundamentalvision/BEVFormer","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"CAPE","metrics":{"Future Frame":"false","NDS":"62.8"},"uses_additional_data":false,"paper_date":"2023-03-17","paper":"/paper/cape-camera-view-position-embedding-for-multi","paper_url":"https://arxiv.org/abs/2303.10209v1","paper_title":"CAPE: Camera View Position Embedding for Multi-View 3D Object Detection","code":"https://github.com/PaddlePaddle/Paddle3D","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"SA-BEV","metrics":{"Future Frame":"false","NDS":"62.4"},"uses_additional_data":false,"paper_date":"2023-07-21","paper":"/paper/sa-bev-generating-semantic-aware-bird-s-eye","paper_url":"https://arxiv.org/abs/2307.11477v1","paper_title":"SA-BEV: Generating Semantic-Aware Bird's-Eye-View Feature for Multi-view 3D Object Detection","code":"https://github.com/mengtan00/sa-bev","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"SOLOFusion-pure","metrics":{"Future Frame":"false","NDS":"61.9"},"uses_additional_data":false,"paper_date":"2022-10-05","paper":"/paper/time-will-tell-new-outlooks-and-a-baseline","paper_url":"https://arxiv.org/abs/2210.02443v1","paper_title":"Time Will Tell: New Outlooks and A Baseline for Temporal Multi-View 3D Object Detection","code":"https://github.com/divadi/solofusion","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":4,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"BEVStereo","metrics":{"Future Frame":"false","NDS":"61.0"},"uses_additional_data":false,"paper_date":"2022-09-21","paper":"/paper/bevstereo-enhancing-depth-estimation-in-multi","paper_url":"https://arxiv.org/abs/2209.10248v1","paper_title":"BEVStereo: Enhancing Depth Estimation in Multi-view 3D Object Detection with Dynamic Temporal Stereo","code":"https://github.com/megvii-basedetection/bevdepth","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":5,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"BEVDepth-pure","metrics":{"Future Frame":"false","NDS":"60.9"},"uses_additional_data":false,"paper_date":"2022-06-21","paper":"/paper/bevdepth-acquisition-of-reliable-depth-for","paper_url":"https://arxiv.org/abs/2206.10092v2","paper_title":"BEVDepth: Acquisition of Reliable Depth for Multi-view 3D Object Detection","code":"https://github.com/megvii-basedetection/bevdepth","n_code_links":2,"syntology":null},{"rank_in_archive_order":14,"model":"SeaBird","metrics":{"Future Frame":"false","NDS":"59.7"},"uses_additional_data":false,"paper_date":"2024-03-29","paper":"/paper/seabird-segmentation-in-bird-s-view-with-dice","paper_url":"https://arxiv.org/abs/2403.20318v1","paper_title":"SeaBird: Segmentation in Bird's View with Dice Loss Improves Monocular 3D Detection of Large Objects","code":"https://github.com/abhi1kumar/seabird","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"BEVDistill","metrics":{"Future Frame":"false","NDS":"59.4"},"uses_additional_data":false,"paper_date":"2022-11-17","paper":"/paper/bevdistill-cross-modal-bev-distillation-for","paper_url":"https://arxiv.org/abs/2211.09386v1","paper_title":"BEVDistill: Cross-Modal BEV Distillation for Multi-View 3D Object Detection","code":"https://github.com/zehuichen123/bevdistill","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"PETRv2-pure","metrics":{"Future Frame":"false","NDS":"59.2"},"uses_additional_data":false,"paper_date":"2022-06-02","paper":"/paper/petrv2-a-unified-framework-for-3d-perception","paper_url":"https://arxiv.org/abs/2206.01256v3","paper_title":"PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images","code":"https://github.com/megvii-research/petr","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":17,"model":"PolarFormer","metrics":{"Future Frame":"false","NDS":"57.2"},"uses_additional_data":false,"paper_date":"2022-06-30","paper":"/paper/polarformer-multi-camera-3d-object-detection","paper_url":"https://arxiv.org/abs/2206.15398v6","paper_title":"PolarFormer: Multi-camera 3D Object Detection with Polar Transformer","code":"https://github.com/fudan-zvg/polarformer","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":6,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":18,"model":"BEVDet4D","metrics":{"Future Frame":"false","NDS":"56.9"},"uses_additional_data":false,"paper_date":"2022-03-31","paper":"/paper/bevdet4d-exploit-temporal-cues-in-multi","paper_url":"https://arxiv.org/abs/2203.17054v3","paper_title":"BEVDet4D: Exploit Temporal Cues in Multi-camera 3D Object Detection","code":"https://github.com/HuangJunJie2017/BEVDet","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"BEVFormer","metrics":{"Future Frame":"false","NDS":"56.9"},"uses_additional_data":false,"paper_date":"2022-03-31","paper":"/paper/bevformer-learning-bird-s-eye-view","paper_url":"https://arxiv.org/abs/2203.17270v2","paper_title":"BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers","code":"https://github.com/fundamentalvision/BEVFormer","n_code_links":3,"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 6,264 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":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"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":11,"rows_with_any_sample_ran":7,"distinct_papers_with_graph_line":11,"distinct_papers_with_any_sample_ran":7,"samples_over_distinct_papers":{"n_ran":28,"n_unverified":23,"n_samples":51,"n_pointer_only_licence":19,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":28,"n_unverified":23,"n_samples":51,"n_pointer_only_licence":19,"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"}}}