{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/r-c3d-region-convolutional-3d-network-for","title":"R-C3D: Region Convolutional 3D Network for Temporal Activity Detection","arxiv_id":"1703.07814","date":"2017-03-22","proceeding":"ICCV 2017 10","authors":["Huijuan Xu","Abir Das","Kate Saenko"],"abstract":"We address the problem of activity detection in continuous, untrimmed video\nstreams. This is a difficult task that requires extracting meaningful\nspatio-temporal features to capture activities, accurately localizing the start\nand end times of each activity. We introduce a new model, Region Convolutional\n3D Network (R-C3D), which encodes the video streams using a three-dimensional\nfully convolutional network, then generates candidate temporal regions\ncontaining activities, and finally classifies selected regions into specific\nactivities. Computation is saved due to the sharing of convolutional features\nbetween the proposal and the classification pipelines. The entire model is\ntrained end-to-end with jointly optimized localization and classification\nlosses. R-C3D is faster than existing methods (569 frames per second on a\nsingle Titan X Maxwell GPU) and achieves state-of-the-art results on THUMOS'14.\nWe further demonstrate that our model is a general activity detection framework\nthat does not rely on assumptions about particular dataset properties by\nevaluating our approach on ActivityNet and Charades. Our code is available at\nhttp://ai.bu.edu/r-c3d/.","url_abs":"http://arxiv.org/abs/1703.07814v2","url_pdf":"http://arxiv.org/pdf/1703.07814v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"r-c3d-region-convolutional-3d-network-for","repo_url":"https://github.com/VisionLearningGroup/R-C3D","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"r-c3d-region-convolutional-3d-network-for","repo_url":"https://github.com/2023-MindSpore-4/Code8/tree/main/C3D","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"r-c3d-region-convolutional-3d-network-for","repo_url":"https://github.com/MindSpore-paper-code-3/code6/tree/main/C3D","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"action-detection","task_name":"Action Detection"},{"task_slug":"action-recognition-in-videos-2","task_name":"Action Recognition In Videos"},{"task_slug":"activity-detection","task_name":"Activity Detection"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/action-detection-on-charades","task":"Action Detection","dataset":"Charades","model":"R-C3D","rank_in_archive_order":15,"of":16,"metrics":{"mAP":"12.4"},"uses_additional_data":false},{"leaderboard":"/sota/action-recognition-in-videos-on-thumos14-1","task":"Action Recognition In Videos","dataset":"THUMOS’14","model":"Single-stream R-C3D (two-way buffer)","rank_in_archive_order":1,"of":2,"metrics":{"mAP@0.1":"54.5","mAP@0.2":"51.5","mAP@0.3":"44.8","mAP@0.4":"35.6","mAP@0.5":"28.9"},"uses_additional_data":false},{"leaderboard":"/sota/action-recognition-in-videos-on-thumos14-1","task":"Action Recognition In Videos","dataset":"THUMOS’14","model":"Single-stream R-C3D (one-way buffer)","rank_in_archive_order":2,"of":2,"metrics":{"mAP@0.1":"51.6","mAP@0.2":"49.2","mAP@0.3":"42.8","mAP@0.4":"33.4","mAP@0.5":"27.0"},"uses_additional_data":false},{"leaderboard":"/sota/temporal-action-localization-on-thumos14","task":"Temporal Action Localization","dataset":"THUMOS’14","model":"R-C3D","rank_in_archive_order":35,"of":42,"metrics":{"mAP IOU@0.1":"54.5","mAP IOU@0.2":"51.5","mAP IOU@0.3":"44.8","mAP IOU@0.4":"35.6","mAP IOU@0.5":"28.9"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1703.07814","atlas_url":"https://app.syntology.ai/?focus=1703.07814","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1703.07814"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/VisionLearningGroup/R-C3D","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2023-MindSpore-4/Code8/tree/main/C3D","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MindSpore-paper-code-3/code6/tree/main/C3D","reach":null}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"0087afa9a6463790","entry":"py_cpu_nms","repo":"VisionLearningGroup/R-C3D","repo_kind":"official","path":"lib/nms/py_cpu_nms.py","file_url":"https://github.com/VisionLearningGroup/R-C3D/blob/HEAD/lib/nms/py_cpu_nms.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0087afa9a6463790"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}