{"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/flashvtg-feature-layering-and-adaptive-score","title":"FlashVTG: Feature Layering and Adaptive Score Handling Network for Video Temporal Grounding","arxiv_id":"2412.13441","date":"2024-12-18","proceeding":null,"authors":["Zhuo Cao","Bingqing Zhang","Heming Du","Xin Yu","Xue Li","Sen Wang"],"abstract":"Text-guided Video Temporal Grounding (VTG) aims to localize relevant segments in untrimmed videos based on textual descriptions, encompassing two subtasks: Moment Retrieval (MR) and Highlight Detection (HD). Although previous typical methods have achieved commendable results, it is still challenging to retrieve short video moments. This is primarily due to the reliance on sparse and limited decoder queries, which significantly constrain the accuracy of predictions. Furthermore, suboptimal outcomes often arise because previous methods rank predictions based on isolated predictions, neglecting the broader video context. To tackle these issues, we introduce FlashVTG, a framework featuring a Temporal Feature Layering (TFL) module and an Adaptive Score Refinement (ASR) module. The TFL module replaces the traditional decoder structure to capture nuanced video content variations across multiple temporal scales, while the ASR module improves prediction ranking by integrating context from adjacent moments and multi-temporal-scale features. Extensive experiments demonstrate that FlashVTG achieves state-of-the-art performance on four widely adopted datasets in both MR and HD. Specifically, on the QVHighlights dataset, it boosts mAP by 5.8% for MR and 3.3% for HD. For short-moment retrieval, FlashVTG increases mAP to 125% of previous SOTA performance. All these improvements are made without adding training burdens, underscoring its effectiveness. Our code is available at https://github.com/Zhuo-Cao/FlashVTG.","url_abs":"https://arxiv.org/abs/2412.13441v1","url_pdf":"https://arxiv.org/pdf/2412.13441v1.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":"flashvtg-feature-layering-and-adaptive-score","repo_url":"https://github.com/zhuo-cao/flashvtg","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"highlight-detection","task_name":"Highlight Detection"},{"task_slug":"moment-retrieval","task_name":"Moment Retrieval"},{"task_slug":"natural-language-moment-retrieval","task_name":"Natural Language Moment Retrieval"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"video-understanding","task_name":"Video Understanding"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/highlight-detection-on-qvhighlights","task":"Highlight Detection","dataset":"QVHighlights","model":"FlashVTG","rank_in_archive_order":2,"of":21,"metrics":{"Hit@1":"71.01","mAP":"44.09"},"uses_additional_data":false},{"leaderboard":"/sota/highlight-detection-on-tvsum","task":"Highlight Detection","dataset":"TvSum","model":"FlashVTG","rank_in_archive_order":1,"of":7,"metrics":{"mAP":"88"},"uses_additional_data":false},{"leaderboard":"/sota/highlight-detection-on-youtube-highlights","task":"Highlight Detection","dataset":"YouTube Highlights","model":"FlashVTG","rank_in_archive_order":5,"of":7,"metrics":{"mAP":"75.4"},"uses_additional_data":false},{"leaderboard":"/sota/moment-retrieval-on-charades-sta","task":"Moment Retrieval","dataset":"Charades-STA","model":"FlashVTG","rank_in_archive_order":3,"of":25,"metrics":{"R@1 IoU=0.5":"70.32","R@1 IoU=0.7":"49.87"},"uses_additional_data":false},{"leaderboard":"/sota/moment-retrieval-on-qvhighlights","task":"Moment Retrieval","dataset":"QVHighlights","model":"FlashVTG","rank_in_archive_order":4,"of":32,"metrics":{"R@1 IoU=0.5":"70.69","R@1 IoU=0.7":"53.96","mAP":"52.00","mAP@0.5":"72.33","mAP@0.75":"53.85"},"uses_additional_data":false},{"leaderboard":"/sota/natural-language-moment-retrieval-on-tacos","task":"Natural Language Moment Retrieval","dataset":"TACoS","model":"FlashVTG","rank_in_archive_order":6,"of":13,"metrics":{"R@1,IoU=0.3":"53.71","R@1,IoU=0.5":"41.76","R@1,IoU=0.7":"24.74","mIoU":"37.61"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2412.13441","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2412.13441"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/zhuo-cao/flashvtg","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":1,"ran_violates":1,"unverified":11},"by_repo_kind":{"official":{"samples":13,"ran":2,"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":13,"samples":[{"code_sha256_prefix":"6897880b6ce14c94","entry":"inverse_sigmoid","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/transformer.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/transformer.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6897880b6ce14c94"}},{"code_sha256_prefix":"b7639c2e8b83837b","entry":"span_cxw_to_xx","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/span_utils.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/span_utils.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b7639c2e8b83837b"}},{"code_sha256_prefix":"081c8abd57d082fd","entry":"accuracy","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/misc.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/misc.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"081c8abd57d082fd"}},{"code_sha256_prefix":"1affbf903fa3204b","entry":"build_position_encoding","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/position_encoding.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/position_encoding.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1affbf903fa3204b"}},{"code_sha256_prefix":"2359afb511e0c62c","entry":"build_transformer","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/transformer.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/transformer.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"2359afb511e0c62c"}},{"code_sha256_prefix":"6d07870656ce939e","entry":"element_wise_list_equal","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/model.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6d07870656ce939e"}},{"code_sha256_prefix":"f2157348a4ae2f51","entry":"find_nth","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/model.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f2157348a4ae2f51"}},{"code_sha256_prefix":"84a83d94394b10af","entry":"gen_sineembed_for_position","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/transformer.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/transformer.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"84a83d94394b10af"}},{"code_sha256_prefix":"a82091666b9db8cf","entry":"multi_head_attention_forward","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/attention.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/attention.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a82091666b9db8cf"}},{"code_sha256_prefix":"5d7fc668e3b608cf","entry":"multi_head_attention_forward","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/crossattention.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/crossattention.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5d7fc668e3b608cf"}},{"code_sha256_prefix":"ae584c65efd4133c","entry":"prepare_batch_inputs","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/start_end_dataset.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/start_end_dataset.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ae584c65efd4133c"}},{"code_sha256_prefix":"7114f025044af9b4","entry":"span_xx_to_cxw","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/span_utils.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/span_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7114f025044af9b4"}},{"code_sha256_prefix":"e4ae5582a2533cdb","entry":"temporal_iou","repo":"zhuo-cao/flashvtg","repo_kind":"official","path":"FlashVTG/span_utils.py","file_url":"https://github.com/zhuo-cao/flashvtg/blob/HEAD/FlashVTG/span_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e4ae5582a2533cdb"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}