{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/task/action-detection/papers/2","list_of":"/task/action-detection","task":"Action Detection","archive":{"snapshot":"2025-07-28"},"key_notes":{"n_ran_checked":"legacy name, kept unchanged so existing readers do not break: it counts the samples that ran with no instrument failure (honoured, violated, and ran with no contract checked); it does not mean a contract was checked, and the pages print it as 'K with no instrument failure', not 'K checked'","n_constructed":"a sub-count of the samples that ran, never subtracted from them and never a failure: an executed sample whose run returned an instance of its own class (fixture_out_type equals the entry name): the run built an object and did not compute a result (Syntology's RAN record, counts.constructed)"},"syntology_read_at":"2026-09-28T10:30:06+00:00","order":"archive","order_definition":"repositories listed in the archive (most first), then date (newest first), then slug","page":2,"pages_in_order":9,"rows_per_page":100,"rows":[101,200],"of":817,"counts":{"archive_papers_tagged":817,"with_a_code_link":277,"where_syntology_ran_a_sample":53,"not_listed_spam_title":0,"listed":817,"listed_where_code_ran":53,"where_syntology_ran_a_sample_split":{"with_a_run_with_no_instrument_failure":46,"every_run_a_failure_of_syntologys_instrument":7,"listed_with_a_run_with_no_instrument_failure":46,"listed_every_run_a_failure_of_syntologys_instrument":7,"filter":{"states":["a run with no instrument failure","any run, instrument failures included"],"default":"a run with no instrument failure","note":"on the 'only where code ran' pages the default hides, in the browser, the rows where every run was a failure of Syntology's instrument; the second state shows them again. Rows are hidden, never re-ordered; these twins list every row"}},"definition":"distinct papers the archive tags; 'where Syntology ran a sample' counts papers with at least one harvested sample that ran, which is not a correctness claim"},"first_page":"/task/action-detection","prev":"/task/action-detection","next":"/task/action-detection/papers/3","papers":[{"url":"/paper/adapting-short-term-transformers-for-action","slug":"adapting-short-term-transformers-for-action","title":"Adapting Short-Term Transformers for Action Detection in Untrimmed Videos","date":"2023-12-04","arxiv_id":"2312.01897","repositories_listed":1,"syntology":null},{"url":"/paper/centre-stage-centricity-based-audio-visual","slug":"centre-stage-centricity-based-audio-visual","title":"Centre Stage: Centricity-based Audio-Visual Temporal Action Detection","date":"2023-11-28","arxiv_id":"2311.16446","repositories_listed":1,"syntology":null},{"url":"/paper/chimpact-a-longitudinal-dataset-for-1","slug":"chimpact-a-longitudinal-dataset-for-1","title":"ChimpACT: A Longitudinal Dataset for Understanding Chimpanzee Behaviors","date":"2023-10-25","arxiv_id":"2310.16447","repositories_listed":1,"syntology":{"n":2,"n_ran":2,"n_constructed":0,"n_ran_checked":2,"n_instrument":0,"n_unverified":0,"n_honours":0,"n_violates":0,"n_no_contract":2,"n_pointer_only":2,"phrase":"2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified","sample_list":"/paper/chimpact-a-longitudinal-dataset-for-1#ran","syntology_url":"https://syntology.ai/paper/2310.16447","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.16447"}},"official":{"repos":["shirleymaxx/chimpact"],"state":"official (archive's flag): 2 ran","n_ran":2,"n_constructed":0,"n_ran_no_instrument_failure":2,"n_unverified":0,"ran_from_kinds":["official"]}}},{"url":"/paper/boundary-discretization-and-reliable","slug":"boundary-discretization-and-reliable","title":"Boundary Discretization and Reliable Classification Network for Temporal Action Detection","date":"2023-10-10","arxiv_id":"2310.06403","repositories_listed":1,"syntology":null},{"url":"/paper/enigma-51-towards-a-fine-grained","slug":"enigma-51-towards-a-fine-grained","title":"ENIGMA-51: Towards a Fine-Grained Understanding of Human-Object Interactions in Industrial Scenarios","date":"2023-09-26","arxiv_id":"2309.14809","repositories_listed":1,"syntology":null},{"url":"/paper/comedian-self-supervised-learning-and","slug":"comedian-self-supervised-learning-and","title":"COMEDIAN: Self-Supervised Learning and Knowledge Distillation for Action Spotting using Transformers","date":"2023-09-03","arxiv_id":"2309.01270","repositories_listed":1,"syntology":null},{"url":"/paper/progression-guided-temporal-action-detection","slug":"progression-guided-temporal-action-detection","title":"Progression-Guided Temporal Action Detection in Videos","date":"2023-08-18","arxiv_id":"2308.09268","repositories_listed":1,"syntology":null},{"url":"/paper/memory-and-anticipation-transformer-for","slug":"memory-and-anticipation-transformer-for","title":"Memory-and-Anticipation Transformer for Online Action Understanding","date":"2023-08-15","arxiv_id":"2308.07893","repositories_listed":1,"syntology":{"n":1,"n_ran":0,"n_constructed":0,"n_ran_checked":0,"n_instrument":0,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":0,"n_pointer_only":0,"phrase":"0 ran · 1 unverified","sample_list":"/paper/memory-and-anticipation-transformer-for#ran","syntology_url":"https://syntology.ai/paper/2308.07893","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2308.07893"}},"official":{"repos":["echo0125/memory-and-anticipation-transformer"],"state":"official: harvested, nothing ran","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":1,"ran_from_kinds":[]}}},{"url":"/paper/integrating-emotion-recognition-with-speech","slug":"integrating-emotion-recognition-with-speech","title":"Integrating Emotion Recognition with Speech Recognition and Speaker Diarisation for Conversations","date":"2023-08-14","arxiv_id":"2308.07145","repositories_listed":1,"syntology":null},{"url":"/paper/long-term-conversation-analysis-exploring","slug":"long-term-conversation-analysis-exploring","title":"Long-term Conversation Analysis: Exploring Utility and Privacy","date":"2023-06-28","arxiv_id":"2306.16071","repositories_listed":1,"syntology":null},{"url":"/paper/e2e-load-end-to-end-long-form-online-action","slug":"e2e-load-end-to-end-long-form-online-action","title":"E2E-LOAD: End-to-End Long-form Online Action Detection","date":"2023-06-13","arxiv_id":"2306.07703","repositories_listed":1,"syntology":{"n":16,"n_ran":12,"n_constructed":0,"n_ran_checked":12,"n_instrument":0,"n_unverified":4,"n_honours":0,"n_violates":0,"n_no_contract":12,"n_pointer_only":16,"phrase":"12 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified","sample_list":"/paper/e2e-load-end-to-end-long-form-online-action#ran","syntology_url":"https://syntology.ai/paper/2306.07703","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2306.07703"}},"official":{"repos":["sqiangcao99/e2e-load"],"state":"official (archive's flag): 12 ran","n_ran":12,"n_constructed":0,"n_ran_no_instrument_failure":12,"n_unverified":4,"ran_from_kinds":["official"]}}},{"url":"/paper/funasr-a-fundamental-end-to-end-speech","slug":"funasr-a-fundamental-end-to-end-speech","title":"FunASR: A Fundamental End-to-End Speech Recognition Toolkit","date":"2023-05-18","arxiv_id":"2305.11013","repositories_listed":1,"syntology":null},{"url":"/paper/wear-a-multimodal-dataset-for-wearable-and","slug":"wear-a-multimodal-dataset-for-wearable-and","title":"WEAR: An Outdoor Sports Dataset for Wearable and Egocentric Activity Recognition","date":"2023-04-11","arxiv_id":"2304.05088","repositories_listed":1,"syntology":null},{"url":"/paper/interaction-aware-prompting-for-zero-shot","slug":"interaction-aware-prompting-for-zero-shot","title":"Interaction-Aware Prompting for Zero-Shot Spatio-Temporal Action Detection","date":"2023-04-10","arxiv_id":"2304.04688","repositories_listed":1,"syntology":null},{"url":"/paper/boundary-denoising-for-video-activity","slug":"boundary-denoising-for-video-activity","title":"Boundary-Denoising for Video Activity Localization","date":"2023-04-06","arxiv_id":"2304.02934","repositories_listed":1,"syntology":null},{"url":"/paper/evaluation-of-noise-reduction-methods-for","slug":"evaluation-of-noise-reduction-methods-for","title":"Evaluation of Noise Reduction Methods for Sentence Recognition by Sinhala Speaking Listeners","date":"2023-03-31","arxiv_id":"2303.17829","repositories_listed":1,"syntology":null},{"url":"/paper/difftad-temporal-action-detection-with","slug":"difftad-temporal-action-detection-with","title":"DiffTAD: Temporal Action Detection with Proposal Denoising Diffusion","date":"2023-03-27","arxiv_id":"2303.14863","repositories_listed":1,"syntology":null},{"url":"/paper/tridet-temporal-action-detection-with","slug":"tridet-temporal-action-detection-with","title":"TriDet: Temporal Action Detection with Relative Boundary Modeling","date":"2023-03-13","arxiv_id":"2303.07347","repositories_listed":1,"syntology":{"n":15,"n_ran":12,"n_constructed":0,"n_ran_checked":10,"n_instrument":2,"n_unverified":3,"n_honours":1,"n_violates":0,"n_no_contract":9,"n_pointer_only":12,"phrase":"12 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified","sample_list":"/paper/tridet-temporal-action-detection-with#ran","syntology_url":"https://syntology.ai/paper/2303.07347","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2303.07347"}},"official":{"repos":["dingfengshi/tridet"],"state":"official (archive's flag): 12 ran","n_ran":12,"n_constructed":0,"n_ran_no_instrument_failure":10,"n_unverified":3,"ran_from_kinds":["official"]}}},{"url":"/paper/ts-sep-joint-diarization-and-separation","slug":"ts-sep-joint-diarization-and-separation","title":"TS-SEP: Joint Diarization and Separation Conditioned on Estimated Speaker Embeddings","date":"2023-03-07","arxiv_id":"2303.03849","repositories_listed":1,"syntology":null},{"url":"/paper/minotaur-multi-task-video-grounding-from","slug":"minotaur-multi-task-video-grounding-from","title":"MINOTAUR: Multi-task Video Grounding From Multimodal Queries","date":"2023-02-16","arxiv_id":"2302.08063","repositories_listed":1,"syntology":null},{"url":"/paper/yowov2-a-stronger-yet-efficient-multi-level","slug":"yowov2-a-stronger-yet-efficient-multi-level","title":"YOWOv2: A Stronger yet Efficient Multi-level Detection Framework for Real-time Spatio-temporal Action Detection","date":"2023-02-14","arxiv_id":"2302.06848","repositories_listed":1,"syntology":null},{"url":"/paper/baseline-method-for-the-sport-task-of","slug":"baseline-method-for-the-sport-task-of","title":"Baseline Method for the Sport Task of MediaEval 2022 with 3D CNNs using Attention Mechanisms","date":"2023-02-06","arxiv_id":"2302.02752","repositories_listed":1,"syntology":null},{"url":"/paper/fine-grained-action-detection-with-rgb-and","slug":"fine-grained-action-detection-with-rgb-and","title":"Fine-Grained Action Detection with RGB and Pose Information using Two Stream Convolutional Networks","date":"2023-02-06","arxiv_id":"2302.02755","repositories_listed":1,"syntology":null},{"url":"/paper/sport-task-fine-grained-action-detection-and","slug":"sport-task-fine-grained-action-detection-and","title":"Sport Task: Fine Grained Action Detection and Classification of Table Tennis Strokes from Videos for MediaEval 2022","date":"2023-01-31","arxiv_id":"2301.13576","repositories_listed":1,"syntology":null},{"url":"/paper/miniroad-minimal-rnn-framework-for-online","slug":"miniroad-minimal-rnn-framework-for-online","title":"MiniROAD: Minimal RNN Framework for Online Action Detection","date":"2023-01-01","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/contextual-explainable-video-representation","slug":"contextual-explainable-video-representation","title":"Contextual Explainable Video Representation: Human Perception-based Understanding","date":"2022-12-12","arxiv_id":"2212.06206","repositories_listed":1,"syntology":null},{"url":"/paper/multi-modal-few-shot-temporal-action","slug":"multi-modal-few-shot-temporal-action","title":"Multi-Modal Few-Shot Temporal Action Detection","date":"2022-11-27","arxiv_id":"2211.14905","repositories_listed":1,"syntology":null},{"url":"/paper/post-processing-temporal-action-detection","slug":"post-processing-temporal-action-detection","title":"Post-Processing Temporal Action Detection","date":"2022-11-27","arxiv_id":"2211.14924","repositories_listed":1,"syntology":null},{"url":"/paper/token-turing-machines","slug":"token-turing-machines","title":"Token Turing Machines","date":"2022-11-16","arxiv_id":"2211.09119","repositories_listed":1,"syntology":null},{"url":"/paper/sg-vad-stochastic-gates-based-speech-activity","slug":"sg-vad-stochastic-gates-based-speech-activity","title":"SG-VAD: Stochastic Gates Based Speech Activity Detection","date":"2022-10-28","arxiv_id":"2210.16022","repositories_listed":1,"syntology":null},{"url":"/paper/handwashing-action-detection-system-for-an","slug":"handwashing-action-detection-system-for-an","title":"Handwashing Action Detection System for an Autonomous Social Robot","date":"2022-10-27","arxiv_id":"2210.15804","repositories_listed":1,"syntology":null},{"url":"/paper/multitask-detection-of-speaker-changes","slug":"multitask-detection-of-speaker-changes","title":"Multitask Detection of Speaker Changes, Overlapping Speech and Voice Activity Using wav2vec 2.0","date":"2022-10-26","arxiv_id":"2210.14755","repositories_listed":1,"syntology":null},{"url":"/paper/refining-action-boundaries-for-one-stage","slug":"refining-action-boundaries-for-one-stage","title":"Refining Action Boundaries for One-stage Detection","date":"2022-10-25","arxiv_id":"2210.14284","repositories_listed":1,"syntology":null},{"url":"/paper/brouhaha-multi-task-training-for-voice","slug":"brouhaha-multi-task-training-for-voice","title":"Brouhaha: multi-task training for voice activity detection, speech-to-noise ratio, and C50 room acoustics estimation","date":"2022-10-24","arxiv_id":"2210.13248","repositories_listed":1,"syntology":null},{"url":"/paper/holistic-interaction-transformer-network-for","slug":"holistic-interaction-transformer-network-for","title":"Holistic Interaction Transformer Network for Action Detection","date":"2022-10-23","arxiv_id":"2210.12686","repositories_listed":1,"syntology":null},{"url":"/paper/pointtad-multi-label-temporal-action","slug":"pointtad-multi-label-temporal-action","title":"PointTAD: Multi-Label Temporal Action Detection with Learnable Query Points","date":"2022-10-20","arxiv_id":"2210.11035","repositories_listed":1,"syntology":{"n":8,"n_ran":6,"n_constructed":1,"n_ran_checked":6,"n_instrument":0,"n_unverified":2,"n_honours":0,"n_violates":0,"n_no_contract":6,"n_pointer_only":3,"phrase":"6 ran (of which 1 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified","sample_list":"/paper/pointtad-multi-label-temporal-action#ran","syntology_url":"https://syntology.ai/paper/2210.11035","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2210.11035"}},"official":{"repos":["mcg-nju/pointtad"],"state":"official (archive's flag): 1 ran","n_ran":1,"n_constructed":1,"n_ran_no_instrument_failure":1,"n_unverified":0,"ran_from_kinds":["found_in_text","official"]}}},{"url":"/paper/yowo-plus-an-incremental-improvement","slug":"yowo-plus-an-incremental-improvement","title":"YOWO-Plus: An Incremental Improvement","date":"2022-10-20","arxiv_id":"2210.11219","repositories_listed":1,"syntology":null},{"url":"/paper/aoe-net-entities-interactions-modeling-with","slug":"aoe-net-entities-interactions-modeling-with","title":"AOE-Net: Entities Interactions Modeling with Adaptive Attention Mechanism for Temporal Action Proposals Generation","date":"2022-10-05","arxiv_id":"2210.02578","repositories_listed":1,"syntology":null},{"url":"/paper/exploiting-instance-based-mixed-sampling-via","slug":"exploiting-instance-based-mixed-sampling-via","title":"Exploiting Instance-based Mixed Sampling via Auxiliary Source Domain Supervision for Domain-adaptive Action Detection","date":"2022-09-28","arxiv_id":"2209.15439","repositories_listed":1,"syntology":null},{"url":"/paper/radacs-towards-higher-order-reasoning-using","slug":"radacs-towards-higher-order-reasoning-using","title":"RALACs: Action Recognition in Autonomous Vehicles using Interaction Encoding and Optical Flow","date":"2022-09-28","arxiv_id":"2209.14408","repositories_listed":1,"syntology":null},{"url":"/paper/exploring-modulated-detection-transformer-as","slug":"exploring-modulated-detection-transformer-as","title":"Exploring Modulated Detection Transformer as a Tool for Action Recognition in Videos","date":"2022-09-21","arxiv_id":"2209.10126","repositories_listed":1,"syntology":null},{"url":"/paper/real-time-online-video-detection-with","slug":"real-time-online-video-detection-with","title":"Real-time Online Video Detection with Temporal Smoothing Transformers","date":"2022-09-19","arxiv_id":"2209.09236","repositories_listed":1,"syntology":{"n":4,"n_ran":3,"n_constructed":0,"n_ran_checked":3,"n_instrument":0,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":3,"n_pointer_only":0,"phrase":"3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified","sample_list":"/paper/real-time-online-video-detection-with#ran","syntology_url":"https://syntology.ai/paper/2209.09236","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2209.09236"}},"official":{"repos":["zhaoyue-zephyrus/testra"],"state":"official (archive's flag): 3 ran","n_ran":3,"n_constructed":0,"n_ran_no_instrument_failure":3,"n_unverified":1,"ran_from_kinds":["official"]}}},{"url":"/paper/actor-identified-spatiotemporal-action","slug":"actor-identified-spatiotemporal-action","title":"Actor-identified Spatiotemporal Action Detection --- Detecting Who Is Doing What in Videos","date":"2022-08-27","arxiv_id":"2208.12940","repositories_listed":1,"syntology":null},{"url":"/paper/weakly-supervised-online-action-detection-for","slug":"weakly-supervised-online-action-detection-for","title":"Weakly Supervised Online Action Detection for Infant General Movements","date":"2022-08-07","arxiv_id":"2208.03648","repositories_listed":1,"syntology":null},{"url":"/paper/weakly-supervised-temporal-action-detection","slug":"weakly-supervised-temporal-action-detection","title":"Weakly-Supervised Temporal Action Detection for Fine-Grained Videos with Hierarchical Atomic Actions","date":"2022-07-24","arxiv_id":"2207.11805","repositories_listed":1,"syntology":null},{"url":"/paper/hierarchically-self-supervised-transformer","slug":"hierarchically-self-supervised-transformer","title":"Hierarchically Self-Supervised Transformer for Human Skeleton Representation Learning","date":"2022-07-20","arxiv_id":"2207.09644","repositories_listed":1,"syntology":{"n":10,"n_ran":9,"n_constructed":5,"n_ran_checked":9,"n_instrument":0,"n_unverified":1,"n_honours":2,"n_violates":0,"n_no_contract":7,"n_pointer_only":0,"phrase":"9 ran (of which 5 constructed an object rather than computing a result; 9 with no instrument failure: 2 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified","sample_list":"/paper/hierarchically-self-supervised-transformer#ran","syntology_url":"https://syntology.ai/paper/2207.09644","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2207.09644"}},"official":{"repos":["yuxiaochen1103/Hi-TRS"],"state":"official (archive's flag): 9 ran","n_ran":9,"n_constructed":5,"n_ran_no_instrument_failure":9,"n_unverified":1,"ran_from_kinds":["official"]}}},{"url":"/paper/zero-shot-temporal-action-detection-via","slug":"zero-shot-temporal-action-detection-via","title":"Zero-Shot Temporal Action Detection via Vision-Language Prompting","date":"2022-07-17","arxiv_id":"2207.08184","repositories_listed":1,"syntology":null},{"url":"/paper/react-temporal-action-detection-with","slug":"react-temporal-action-detection-with","title":"ReAct: Temporal Action Detection with Relational Queries","date":"2022-07-14","arxiv_id":"2207.07097","repositories_listed":1,"syntology":{"n":12,"n_ran":7,"n_constructed":0,"n_ran_checked":5,"n_instrument":2,"n_unverified":5,"n_honours":0,"n_violates":0,"n_no_contract":5,"n_pointer_only":2,"phrase":"7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified","sample_list":"/paper/react-temporal-action-detection-with#ran","syntology_url":"https://syntology.ai/paper/2207.07097","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2207.07097"}},"official":{"repos":["sssste/react"],"state":"official (archive's flag): 7 ran","n_ran":7,"n_constructed":0,"n_ran_no_instrument_failure":5,"n_unverified":5,"ran_from_kinds":["official"]}}},{"url":"/paper/semi-supervised-temporal-action-detection","slug":"semi-supervised-temporal-action-detection","title":"Semi-Supervised Temporal Action Detection with Proposal-Free Masking","date":"2022-07-14","arxiv_id":"2207.07059","repositories_listed":1,"syntology":{"n":4,"n_ran":3,"n_constructed":1,"n_ran_checked":3,"n_instrument":0,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":3,"n_pointer_only":4,"phrase":"3 ran (of which 1 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified","sample_list":"/paper/semi-supervised-temporal-action-detection#ran","syntology_url":"https://syntology.ai/paper/2207.07059","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2207.07059"}},"official":{"repos":["sauradip/spot"],"state":"official (archive's flag): 3 ran","n_ran":3,"n_constructed":1,"n_ran_no_instrument_failure":3,"n_unverified":1,"ran_from_kinds":["official"]}}},{"url":"/paper/mm-alt-a-multimodal-automatic-lyric","slug":"mm-alt-a-multimodal-automatic-lyric","title":"MM-ALT: A Multimodal Automatic Lyric Transcription System","date":"2022-07-13","arxiv_id":"2207.06127","repositories_listed":1,"syntology":null},{"url":"/paper/a-semi-supervised-geometric-driven","slug":"a-semi-supervised-geometric-driven","title":"A semi-supervised methodology for fishing activity detection using the geometry behind the trajectory of multiple vessels","date":"2022-07-12","arxiv_id":"2207.05514","repositories_listed":1,"syntology":null},{"url":"/paper/adversarial-multi-task-deep-learning-for","slug":"adversarial-multi-task-deep-learning-for","title":"Adversarial Multi-Task Deep Learning for Noise-Robust Voice Activity Detection with Low Algorithmic Delay","date":"2022-07-04","arxiv_id":"2207.01691","repositories_listed":1,"syntology":null},{"url":"/paper/unsupervised-voice-activity-detection-by","slug":"unsupervised-voice-activity-detection-by","title":"Unsupervised Voice Activity Detection by Modeling Source and System Information using Zero Frequency Filtering","date":"2022-06-27","arxiv_id":"2206.13420","repositories_listed":1,"syntology":null},{"url":"/paper/pyramid-region-based-slot-attention-network","slug":"pyramid-region-based-slot-attention-network","title":"Pyramid Region-based Slot Attention Network for Temporal Action Proposal Generation","date":"2022-06-21","arxiv_id":"2206.10095","repositories_listed":1,"syntology":null},{"url":"/paper/minimum-efforts-to-build-an-end-to-end","slug":"minimum-efforts-to-build-an-end-to-end","title":"A Simple and Efficient Pipeline to Build an End-to-End Spatial-Temporal Action Detector","date":"2022-06-07","arxiv_id":"2206.03064","repositories_listed":1,"syntology":null},{"url":"/paper/tadml-a-fast-temporal-action-detection-with","slug":"tadml-a-fast-temporal-action-detection-with","title":"TadML: A fast temporal action detection with Mechanics-MLP","date":"2022-06-07","arxiv_id":"2206.02997","repositories_listed":1,"syntology":null},{"url":"/paper/stargazer-a-transformer-based-driver-action","slug":"stargazer-a-transformer-based-driver-action","title":"Stargazer: A transformer-based driver action detection system for intelligent transportation","date":"2022-06-01","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/etad-a-unified-framework-for-efficient","slug":"etad-a-unified-framework-for-efficient","title":"ETAD: Training Action Detection End to End on a Laptop","date":"2022-05-14","arxiv_id":"2205.07134","repositories_listed":1,"syntology":null},{"url":"/paper/estimation-of-reliable-proposal-quality-for","slug":"estimation-of-reliable-proposal-quality-for","title":"Estimation of Reliable Proposal Quality for Temporal Action Detection","date":"2022-04-25","arxiv_id":"2204.11695","repositories_listed":1,"syntology":null},{"url":"/paper/a-multi-person-video-dataset-annotation","slug":"a-multi-person-video-dataset-annotation","title":"A Multi-Person Video Dataset Annotation Method of Spatio-Temporally Actions","date":"2022-04-21","arxiv_id":"2204.10160","repositories_listed":1,"syntology":null},{"url":"/paper/e-2tad-an-energy-efficient-tracking-based","slug":"e-2tad-an-energy-efficient-tracking-based","title":"E^2TAD: An Energy-Efficient Tracking-based Action Detector","date":"2022-04-09","arxiv_id":"2204.04416","repositories_listed":1,"syntology":null},{"url":"/paper/an-empirical-study-of-end-to-end-temporal","slug":"an-empirical-study-of-end-to-end-temporal","title":"An Empirical Study of End-to-End Temporal Action Detection","date":"2022-04-06","arxiv_id":"2204.02932","repositories_listed":1,"syntology":{"n":11,"n_ran":10,"n_constructed":0,"n_ran_checked":9,"n_instrument":1,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":9,"n_pointer_only":11,"phrase":"10 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 0 violated, 9 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified","sample_list":"/paper/an-empirical-study-of-end-to-end-temporal#ran","syntology_url":"https://syntology.ai/paper/2204.02932","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2204.02932"}},"official":{"repos":["xlliu7/E2E-TAD"],"state":"official (archive's flag): 10 ran","n_ran":10,"n_constructed":0,"n_ran_no_instrument_failure":9,"n_unverified":1,"ran_from_kinds":["official"]}}},{"url":"/paper/leveraging-speech-separation-for","slug":"leveraging-speech-separation-for","title":"Low-Latency Speech Separation Guided Diarization for Telephone Conversations","date":"2022-04-05","arxiv_id":"2204.02306","repositories_listed":1,"syntology":null},{"url":"/paper/gan-based-joint-activity-detection-and","slug":"gan-based-joint-activity-detection-and","title":"Gan-Based Joint Activity Detection and Channel Estimation For Grant-free Random Access","date":"2022-04-04","arxiv_id":"2204.01731","repositories_listed":1,"syntology":null},{"url":"/paper/stochastic-backpropagation-a-memory-efficient","slug":"stochastic-backpropagation-a-memory-efficient","title":"Stochastic Backpropagation: A Memory Efficient Strategy for Training Video Models","date":"2022-03-31","arxiv_id":"2203.16755","repositories_listed":1,"syntology":null},{"url":"/paper/speaker-embedding-aware-neural-diarization-a","slug":"speaker-embedding-aware-neural-diarization-a","title":"Speaker Embedding-aware Neural Diarization: an Efficient Framework for Overlapping Speech Diarization in Meeting Scenarios","date":"2022-03-18","arxiv_id":"2203.09767","repositories_listed":1,"syntology":null},{"url":"/paper/abn-agent-aware-boundary-networks-for","slug":"abn-agent-aware-boundary-networks-for","title":"ABN: Agent-Aware Boundary Networks for Temporal Action Proposal Generation","date":"2022-03-16","arxiv_id":"2203.08942","repositories_listed":1,"syntology":null},{"url":"/paper/end-to-end-semi-supervised-learning-for-video","slug":"end-to-end-semi-supervised-learning-for-video","title":"End-to-End Semi-Supervised Learning for Video Action Detection","date":"2022-03-08","arxiv_id":"2203.04251","repositories_listed":1,"syntology":{"n":1,"n_ran":1,"n_constructed":0,"n_ran_checked":1,"n_instrument":0,"n_unverified":0,"n_honours":0,"n_violates":0,"n_no_contract":1,"n_pointer_only":0,"phrase":"1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified","sample_list":"/paper/end-to-end-semi-supervised-learning-for-video#ran","syntology_url":"https://syntology.ai/paper/2203.04251","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2203.04251"}},"official":{"repos":["AKASH2907/End-to-End-Semi-Supervised-Learning-for-Video-Action-Detection"],"state":"official (archive's flag): 1 ran","n_ran":1,"n_constructed":0,"n_ran_no_instrument_failure":1,"n_unverified":0,"ran_from_kinds":["official"]}}},{"url":"/paper/colar-effective-and-efficient-online-action","slug":"colar-effective-and-efficient-online-action","title":"Colar: Effective and Efficient Online Action Detection by Consulting Exemplars","date":"2022-03-02","arxiv_id":"2203.01057","repositories_listed":1,"syntology":{"n":1,"n_ran":0,"n_constructed":0,"n_ran_checked":0,"n_instrument":0,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":0,"n_pointer_only":1,"phrase":"0 ran · 1 unverified","sample_list":"/paper/colar-effective-and-efficient-online-action#ran","syntology_url":"https://syntology.ai/paper/2203.01057","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2203.01057"}},"official":{"repos":["vividle/online-action-detection"],"state":"official: harvested, nothing ran","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":1,"ran_from_kinds":[]}}},{"url":"/paper/hgcn-harmonic-gated-compensation-network-for","slug":"hgcn-harmonic-gated-compensation-network-for","title":"HGCN: Harmonic gated compensation network for speech enhancement","date":"2022-01-30","arxiv_id":"2201.12755","repositories_listed":1,"syntology":null},{"url":"/paper/nas-vad-neural-architecture-search-for-voice","slug":"nas-vad-neural-architecture-search-for-voice","title":"NAS-VAD: Neural Architecture Search for Voice Activity Detection","date":"2022-01-22","arxiv_id":"2201.09032","repositories_listed":1,"syntology":null},{"url":"/paper/continual-transformers-redundancy-free","slug":"continual-transformers-redundancy-free","title":"Continual Transformers: Redundancy-Free Attention for Online Inference","date":"2022-01-17","arxiv_id":"2201.06268","repositories_listed":1,"syntology":null},{"url":"/paper/exploiting-temporal-side-information-in","slug":"exploiting-temporal-side-information-in","title":"Exploiting Temporal Side Information in Massive IoT Connectivity","date":"2022-01-05","arxiv_id":"2201.01485","repositories_listed":1,"syntology":null},{"url":"/paper/spatio-temporal-cnn-baseline-method-for-the","slug":"spatio-temporal-cnn-baseline-method-for-the","title":"Spatio-Temporal CNN baseline method for the Sports Video Task of MediaEval 2021 benchmark","date":"2021-12-16","arxiv_id":"2112.12074","repositories_listed":1,"syntology":null},{"url":"/paper/sports-video-fine-grained-action-detection","slug":"sports-video-fine-grained-action-detection","title":"Sports Video: Fine-Grained Action Detection and Classification of Table Tennis Strokes from Videos for MediaEval 2021","date":"2021-12-16","arxiv_id":"2112.11384","repositories_listed":1,"syntology":null},{"url":"/paper/svip-sequence-verification-for-procedures-in","slug":"svip-sequence-verification-for-procedures-in","title":"SVIP: Sequence VerIfication for Procedures in Videos","date":"2021-12-13","arxiv_id":"2112.06447","repositories_listed":1,"syntology":null},{"url":"/paper/x-vector-based-voice-activity-detection-for-1","slug":"x-vector-based-voice-activity-detection-for-1","title":"X-Vector based voice activity detection for multi-genre broadcast speech-to-text","date":"2021-12-09","arxiv_id":"2112.05016","repositories_listed":1,"syntology":null},{"url":"/paper/dcan-improving-temporal-action-detection-via","slug":"dcan-improving-temporal-action-detection-via","title":"DCAN: Improving Temporal Action Detection via Dual Context Aggregation","date":"2021-12-07","arxiv_id":"2112.03612","repositories_listed":1,"syntology":null},{"url":"/paper/ms-tct-multi-scale-temporal-convtransformer","slug":"ms-tct-multi-scale-temporal-convtransformer","title":"MS-TCT: Multi-Scale Temporal ConvTransformer for Action Detection","date":"2021-12-07","arxiv_id":"2112.03902","repositories_listed":1,"syntology":{"n":1,"n_ran":1,"n_constructed":0,"n_ran_checked":0,"n_instrument":1,"n_unverified":0,"n_honours":0,"n_violates":0,"n_no_contract":0,"n_pointer_only":1,"phrase":"1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified","sample_list":"/paper/ms-tct-multi-scale-temporal-convtransformer#ran","syntology_url":"https://syntology.ai/paper/2112.03902","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2112.03902"}},"official":null}},{"url":"/paper/self-supervised-pretraining-with","slug":"self-supervised-pretraining-with","title":"Weakly-guided Self-supervised Pretraining for Temporal Activity Detection","date":"2021-11-26","arxiv_id":"2111.13675","repositories_listed":1,"syntology":{"n":11,"n_ran":8,"n_constructed":0,"n_ran_checked":8,"n_instrument":0,"n_unverified":3,"n_honours":2,"n_violates":0,"n_no_contract":6,"n_pointer_only":1,"phrase":"8 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 2 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified","sample_list":"/paper/self-supervised-pretraining-with#ran","syntology_url":"https://syntology.ai/paper/2111.13675","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2111.13675"}},"official":{"repos":["kkahatapitiya/SSDet"],"state":"official (archive's flag): 8 ran","n_ran":8,"n_constructed":0,"n_ran_no_instrument_failure":8,"n_unverified":3,"ran_from_kinds":["official"]}}},{"url":"/paper/avaspeech-smad-a-strongly-labelled-speech-and","slug":"avaspeech-smad-a-strongly-labelled-speech-and","title":"AVASpeech-SMAD: A Strongly Labelled Speech and Music Activity Detection Dataset with Label Co-Occurrence","date":"2021-11-02","arxiv_id":"2111.01320","repositories_listed":1,"syntology":null},{"url":"/paper/revisiting-spatio-temporal-layouts-for","slug":"revisiting-spatio-temporal-layouts-for","title":"Revisiting spatio-temporal layouts for compositional action recognition","date":"2021-11-02","arxiv_id":"2111.01936","repositories_listed":1,"syntology":null},{"url":"/paper/aei-actors-environment-interaction-with","slug":"aei-actors-environment-interaction-with","title":"AEI: Actors-Environment Interaction with Adaptive Attention for Temporal Action Proposals Generation","date":"2021-10-21","arxiv_id":"2110.11474","repositories_listed":1,"syntology":null},{"url":"/paper/lstc-boosting-atomic-action-detection-with","slug":"lstc-boosting-atomic-action-detection-with","title":"LSTC: Boosting Atomic Action Detection with Long-Short-Term Context","date":"2021-10-19","arxiv_id":"2110.09819","repositories_listed":1,"syntology":null},{"url":"/paper/object-region-video-transformers-1","slug":"object-region-video-transformers-1","title":"Object-Region Video Transformers","date":"2021-10-13","arxiv_id":"2110.06915","repositories_listed":1,"syntology":{"n":7,"n_ran":6,"n_constructed":0,"n_ran_checked":5,"n_instrument":1,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":5,"n_pointer_only":7,"phrase":"6 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified","sample_list":"/paper/object-region-video-transformers-1#ran","syntology_url":"https://syntology.ai/paper/2110.06915","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2110.06915"}},"official":null}},{"url":"/paper/towards-high-quality-temporal-action","slug":"towards-high-quality-temporal-action","title":"Towards High-Quality Temporal Action Detection with Sparse Proposals","date":"2021-09-18","arxiv_id":"2109.08847","repositories_listed":1,"syntology":null},{"url":"/paper/activity-recognition-for-autism-diagnosis","slug":"activity-recognition-for-autism-diagnosis","title":"Classification of Abnormal Hand Movement for Aiding in Autism Detection: Machine Learning Study","date":"2021-08-18","arxiv_id":"2108.07917","repositories_listed":1,"syntology":null},{"url":"/paper/fine-grained-classroom-activity-detection","slug":"fine-grained-classroom-activity-detection","title":"Fine-Grained Classroom Activity Detection from Audio with Neural Networks","date":"2021-07-29","arxiv_id":"2107.14369","repositories_listed":1,"syntology":null},{"url":"/paper/spot-what-matters-learning-context-using","slug":"spot-what-matters-learning-context-using","title":"Spot What Matters: Learning Context Using Graph Convolutional Networks for Weakly-Supervised Action Detection","date":"2021-07-28","arxiv_id":"2107.13648","repositories_listed":1,"syntology":null},{"url":"/paper/rgb-stream-is-enough-for-temporal-action","slug":"rgb-stream-is-enough-for-temporal-action","title":"RGB Stream Is Enough for Temporal Action Detection","date":"2021-07-09","arxiv_id":"2107.04362","repositories_listed":1,"syntology":null},{"url":"/paper/oadtr-online-action-detection-with","slug":"oadtr-online-action-detection-with","title":"OadTR: Online Action Detection with Transformers","date":"2021-06-21","arxiv_id":"2106.11149","repositories_listed":1,"syntology":{"n":5,"n_ran":5,"n_constructed":4,"n_ran_checked":5,"n_instrument":0,"n_unverified":0,"n_honours":0,"n_violates":0,"n_no_contract":5,"n_pointer_only":0,"phrase":"5 ran (of which 4 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified","sample_list":"/paper/oadtr-online-action-detection-with#ran","syntology_url":"https://syntology.ai/paper/2106.11149","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2106.11149"}},"official":{"repos":["wangxiang1230/OadTR"],"state":"official (archive's flag): 5 ran","n_ran":5,"n_constructed":4,"n_ran_no_instrument_failure":5,"n_unverified":0,"ran_from_kinds":["official"]}}},{"url":"/paper/proposal-relation-network-for-temporal-action","slug":"proposal-relation-network-for-temporal-action","title":"Proposal Relation Network for Temporal Action Detection","date":"2021-06-20","arxiv_id":"2106.11812","repositories_listed":1,"syntology":null},{"url":"/paper/end-to-end-temporal-action-detection-with","slug":"end-to-end-temporal-action-detection-with","title":"End-to-end Temporal Action Detection with Transformer","date":"2021-06-18","arxiv_id":"2106.10271","repositories_listed":1,"syntology":{"n":3,"n_ran":2,"n_constructed":0,"n_ran_checked":0,"n_instrument":2,"n_unverified":1,"n_honours":0,"n_violates":0,"n_no_contract":0,"n_pointer_only":0,"phrase":"2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified","sample_list":"/paper/end-to-end-temporal-action-detection-with#ran","syntology_url":"https://syntology.ai/paper/2106.10271","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2106.10271"}},"official":{"repos":["xlliu7/TadTR"],"state":"official (archive's flag): 2 ran","n_ran":2,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":1,"ran_from_kinds":["official"]}}},{"url":"/paper/modist-motion-distillation-for-self","slug":"modist-motion-distillation-for-self","title":"MaCLR: Motion-aware Contrastive Learning of Representations for Videos","date":"2021-06-17","arxiv_id":"2106.09703","repositories_listed":1,"syntology":{"n":6,"n_ran":4,"n_constructed":0,"n_ran_checked":4,"n_instrument":0,"n_unverified":2,"n_honours":0,"n_violates":0,"n_no_contract":4,"n_pointer_only":3,"phrase":"4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified","sample_list":"/paper/modist-motion-distillation-for-self#ran","syntology_url":"https://syntology.ai/paper/2106.09703","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2106.09703"}},"official":{"repos":["amazon-science/self-supervised-maclr"],"state":"official (archive's flag): 4 ran","n_ran":4,"n_constructed":0,"n_ran_no_instrument_failure":4,"n_unverified":2,"ran_from_kinds":["official"]}}},{"url":"/paper/a-stronger-baseline-for-ego-centric-action","slug":"a-stronger-baseline-for-ego-centric-action","title":"A Stronger Baseline for Ego-Centric Action Detection","date":"2021-06-13","arxiv_id":"2106.06942","repositories_listed":1,"syntology":null},{"url":"/paper/wase-learning-when-to-attend-for-speaker","slug":"wase-learning-when-to-attend-for-speaker","title":"WASE: Learning When to Attend for Speaker Extraction in Cocktail Party Environments","date":"2021-06-13","arxiv_id":"2106.07016","repositories_listed":1,"syntology":null},{"url":"/paper/st-hoi-a-spatial-temporal-baseline-for-human","slug":"st-hoi-a-spatial-temporal-baseline-for-human","title":"ST-HOI: A Spatial-Temporal Baseline for Human-Object Interaction Detection in Videos","date":"2021-05-25","arxiv_id":"2105.11731","repositories_listed":1,"syntology":{"n":4,"n_ran":2,"n_constructed":0,"n_ran_checked":2,"n_instrument":0,"n_unverified":2,"n_honours":0,"n_violates":0,"n_no_contract":2,"n_pointer_only":4,"phrase":"2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified","sample_list":"/paper/st-hoi-a-spatial-temporal-baseline-for-human#ran","syntology_url":"https://syntology.ai/paper/2105.11731","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2105.11731"}},"official":{"repos":["coldmanck/VidHOI"],"state":"official (archive's flag): 2 ran","n_ran":2,"n_constructed":0,"n_ran_no_instrument_failure":2,"n_unverified":2,"ran_from_kinds":["official"]}}},{"url":"/paper/fineaction-a-fined-video-dataset-for-temporal","slug":"fineaction-a-fined-video-dataset-for-temporal","title":"FineAction: A Fine-Grained Video Dataset for Temporal Action Localization","date":"2021-05-24","arxiv_id":"2105.11107","repositories_listed":1,"syntology":null},{"url":"/paper/multisports-a-multi-person-video-dataset-of","slug":"multisports-a-multi-person-video-dataset-of","title":"MultiSports: A Multi-Person Video Dataset of Spatio-Temporally Localized Sports Actions","date":"2021-05-16","arxiv_id":"2105.07404","repositories_listed":1,"syntology":null},{"url":"/paper/coupling-intent-and-action-for-pedestrian","slug":"coupling-intent-and-action-for-pedestrian","title":"Coupling Intent and Action for Pedestrian Crossing Behavior Prediction","date":"2021-05-10","arxiv_id":"2105.04133","repositories_listed":1,"syntology":null}],"record_sha256":"829b3ffc0e1a4578533ac76c74d4b05be6e27eaf54daa25ecf4fca13c07f4367","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}