E.M.Ground: A Temporal Grounding Vid-LLM with Holistic Event Perception and Matching
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915775743262720 |
|---|---|
| author | Nie, Jiahao An, Wenbin Zhang, Gongjie Xu, Yicheng Tan, Yap-Peng Kot, Alex C. Lu, Shijian |
| author_facet | Nie, Jiahao An, Wenbin Zhang, Gongjie Xu, Yicheng Tan, Yap-Peng Kot, Alex C. Lu, Shijian |
| contents | Despite recent advances in Video Large Language Models (Vid-LLMs), Temporal Video Grounding (TVG), which aims to precisely localize time segments corresponding to query events, remains a significant challenge. Existing methods often match start and end frames by comparing frame features with two separate tokens, relying heavily on exact timestamps. However, this approach fails to capture the event's semantic continuity and integrity, leading to ambiguities. To address this, we propose E.M.Ground, a novel Vid-LLM for TVG that focuses on holistic and coherent event perception. E.M.Ground introduces three key innovations: (i) a special <event> token that aggregates information from all frames of a query event, preserving semantic continuity for accurate event matching; (ii) Savitzky-Golay smoothing to reduce noise in token-to-frame similarities across timestamps, improving prediction accuracy; (iii) multi-grained frame feature aggregation to enhance matching reliability and temporal understanding, compensating for compression-induced information loss. Extensive experiments on benchmark datasets show that E.M.Ground consistently outperforms state-of-the-art Vid-LLMs by significant margins. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_05215 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | E.M.Ground: A Temporal Grounding Vid-LLM with Holistic Event Perception and Matching Nie, Jiahao An, Wenbin Zhang, Gongjie Xu, Yicheng Tan, Yap-Peng Kot, Alex C. Lu, Shijian Computer Vision and Pattern Recognition Despite recent advances in Video Large Language Models (Vid-LLMs), Temporal Video Grounding (TVG), which aims to precisely localize time segments corresponding to query events, remains a significant challenge. Existing methods often match start and end frames by comparing frame features with two separate tokens, relying heavily on exact timestamps. However, this approach fails to capture the event's semantic continuity and integrity, leading to ambiguities. To address this, we propose E.M.Ground, a novel Vid-LLM for TVG that focuses on holistic and coherent event perception. E.M.Ground introduces three key innovations: (i) a special <event> token that aggregates information from all frames of a query event, preserving semantic continuity for accurate event matching; (ii) Savitzky-Golay smoothing to reduce noise in token-to-frame similarities across timestamps, improving prediction accuracy; (iii) multi-grained frame feature aggregation to enhance matching reliability and temporal understanding, compensating for compression-induced information loss. Extensive experiments on benchmark datasets show that E.M.Ground consistently outperforms state-of-the-art Vid-LLMs by significant margins. |
| title | E.M.Ground: A Temporal Grounding Vid-LLM with Holistic Event Perception and Matching |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.05215 |