Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2502.12917 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929719045259264 |
|---|---|
| author | Wang, Haicheng Ju, Chen Lin, Weixiong Ma, Chaofan Xiao, Shuai Zhang, Ya Wang, Yanfeng |
| author_facet | Wang, Haicheng Ju, Chen Lin, Weixiong Ma, Chaofan Xiao, Shuai Zhang, Ya Wang, Yanfeng |
| contents | Temporal sentence grounding aims to detect event timestamps described by the natural language query from given untrimmed videos. The existing fully-supervised setting achieves great results but requires expensive annotation costs; while the weakly-supervised setting adopts cheap labels but performs poorly. To pursue high performance with less annotation costs, this paper introduces an intermediate partially-supervised setting, i.e., only short-clip is available during training. To make full use of partial labels, we specially design one contrast-unity framework, with the two-stage goal of implicit-explicit progressive grounding. In the implicit stage, we align event-query representations at fine granularity using comprehensive quadruple contrastive learning: event-query gather, event-background separation, intra-cluster compactness and inter-cluster separability. Then, high-quality representations bring acceptable grounding pseudo-labels. In the explicit stage, to explicitly optimize grounding objectives, we train one fully-supervised model using obtained pseudo-labels for grounding refinement and denoising. Extensive experiments and thoroughly ablations on Charades-STA and ActivityNet Captions demonstrate the significance of partial supervision, as well as our superior performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_12917 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Contrast-Unity for Partially-Supervised Temporal Sentence Grounding Wang, Haicheng Ju, Chen Lin, Weixiong Ma, Chaofan Xiao, Shuai Zhang, Ya Wang, Yanfeng Computer Vision and Pattern Recognition Temporal sentence grounding aims to detect event timestamps described by the natural language query from given untrimmed videos. The existing fully-supervised setting achieves great results but requires expensive annotation costs; while the weakly-supervised setting adopts cheap labels but performs poorly. To pursue high performance with less annotation costs, this paper introduces an intermediate partially-supervised setting, i.e., only short-clip is available during training. To make full use of partial labels, we specially design one contrast-unity framework, with the two-stage goal of implicit-explicit progressive grounding. In the implicit stage, we align event-query representations at fine granularity using comprehensive quadruple contrastive learning: event-query gather, event-background separation, intra-cluster compactness and inter-cluster separability. Then, high-quality representations bring acceptable grounding pseudo-labels. In the explicit stage, to explicitly optimize grounding objectives, we train one fully-supervised model using obtained pseudo-labels for grounding refinement and denoising. Extensive experiments and thoroughly ablations on Charades-STA and ActivityNet Captions demonstrate the significance of partial supervision, as well as our superior performance. |
| title | Contrast-Unity for Partially-Supervised Temporal Sentence Grounding |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.12917 |