OSGNet with MLLM Reranking @ Ego4D Episodic Memory Challenge 2026
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916030943592448 |
|---|---|
| author | Feng, Yisen Qu, Leigang Zhang, Haoyu Chu, Qiaohui Liu, Meng Song, Xuemeng Guan, Weili Nie, Liqiang |
| author_facet | Feng, Yisen Qu, Leigang Zhang, Haoyu Chu, Qiaohui Liu, Meng Song, Xuemeng Guan, Weili Nie, Liqiang |
| contents | In this report, we present our champion solutions for the Natural Language Queries and GoalStep tracks of the Ego4D Episodic Memory Challenge at CVPR 2026. Both tracks require accurately localizing temporal segments from long untrimmed egocentric videos. To address these tasks, we propose a reranking-based framework that effectively leverages the strong video-language reasoning capability of multimodal large language model (MLLM) while preserving the efficiency and candidate recall of conventional localization pipelines. Specifically, we first obtain a set of candidate segments from existing localization model OSGNet, and then employ MLLM to select the segment that best matches the given query, thereby refining the final prediction. Ultimately, our method achieved first place in both the Natural Language Queries and GoalStep tracks. Our code can be found at https://github.com/iLearn-Lab/CVPR25-OSGNet. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_20818 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OSGNet with MLLM Reranking @ Ego4D Episodic Memory Challenge 2026 Feng, Yisen Qu, Leigang Zhang, Haoyu Chu, Qiaohui Liu, Meng Song, Xuemeng Guan, Weili Nie, Liqiang Computer Vision and Pattern Recognition In this report, we present our champion solutions for the Natural Language Queries and GoalStep tracks of the Ego4D Episodic Memory Challenge at CVPR 2026. Both tracks require accurately localizing temporal segments from long untrimmed egocentric videos. To address these tasks, we propose a reranking-based framework that effectively leverages the strong video-language reasoning capability of multimodal large language model (MLLM) while preserving the efficiency and candidate recall of conventional localization pipelines. Specifically, we first obtain a set of candidate segments from existing localization model OSGNet, and then employ MLLM to select the segment that best matches the given query, thereby refining the final prediction. Ultimately, our method achieved first place in both the Natural Language Queries and GoalStep tracks. Our code can be found at https://github.com/iLearn-Lab/CVPR25-OSGNet. |
| title | OSGNet with MLLM Reranking @ Ego4D Episodic Memory Challenge 2026 |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.20818 |