Imagine Before Concentration: Diffusion-Guided Registers Enhance Partially Relevant Video Retrieval
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915915860279296 |
|---|---|
| author | Li, Jun Lou, Xuhang Wang, Jinpeng Wang, Yuting Wang, Yaowei Xia, Shu-Tao Chen, Bin |
| author_facet | Li, Jun Lou, Xuhang Wang, Jinpeng Wang, Yuting Wang, Yaowei Xia, Shu-Tao Chen, Bin |
| contents | Partially Relevant Video Retrieval (PRVR) aims to retrieve untrimmed videos based on text queries that describe only partial events. Existing methods suffer from incomplete global contextual perception, struggling with query ambiguity and local noise induced by spurious responses. To address these issues, we propose DreamPRVR, which adopts a coarse-to-fine representation learning paradigm. The model first generates global contextual semantic registers as coarse-grained highlights spanning the entire video and then concentrates on fine-grained similarity optimization for precise cross-modal matching. Concretely, these registers are generated by initializing from the video-centric distribution produced by a probabilistic variational sampler and then iteratively refined via a text-supervised truncated diffusion model. During this process, textual semantic structure learning constructs a well-formed textual latent space, enhancing the reliability of global perception. The registers are then adaptively fused with video tokens through register-augmented Gaussian attention blocks, enabling context-aware feature learning. Extensive experiments show that DreamPRVR outperforms state-of-the-art methods. Code is released at https://github.com/lijun2005/CVPR26-DreamPRVR. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_03653 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Imagine Before Concentration: Diffusion-Guided Registers Enhance Partially Relevant Video Retrieval Li, Jun Lou, Xuhang Wang, Jinpeng Wang, Yuting Wang, Yaowei Xia, Shu-Tao Chen, Bin Computer Vision and Pattern Recognition Information Retrieval Multimedia Partially Relevant Video Retrieval (PRVR) aims to retrieve untrimmed videos based on text queries that describe only partial events. Existing methods suffer from incomplete global contextual perception, struggling with query ambiguity and local noise induced by spurious responses. To address these issues, we propose DreamPRVR, which adopts a coarse-to-fine representation learning paradigm. The model first generates global contextual semantic registers as coarse-grained highlights spanning the entire video and then concentrates on fine-grained similarity optimization for precise cross-modal matching. Concretely, these registers are generated by initializing from the video-centric distribution produced by a probabilistic variational sampler and then iteratively refined via a text-supervised truncated diffusion model. During this process, textual semantic structure learning constructs a well-formed textual latent space, enhancing the reliability of global perception. The registers are then adaptively fused with video tokens through register-augmented Gaussian attention blocks, enabling context-aware feature learning. Extensive experiments show that DreamPRVR outperforms state-of-the-art methods. Code is released at https://github.com/lijun2005/CVPR26-DreamPRVR. |
| title | Imagine Before Concentration: Diffusion-Guided Registers Enhance Partially Relevant Video Retrieval |
| topic | Computer Vision and Pattern Recognition Information Retrieval Multimedia |
| url | https://arxiv.org/abs/2604.03653 |