UATVR: Uncertainty-Adaptive Text-Video Retrieval
About
With the explosive growth of web videos and emerging large-scale vision-language pre-training models, e.g., CLIP, retrieving videos of interest with text instructions has attracted increasing attention. A common practice is to transfer text-video pairs to the same embedding space and craft cross-modal interactions with certain entities in specific granularities for semantic correspondence. Unfortunately, the intrinsic uncertainties of optimal entity combinations in appropriate granularities for cross-modal queries are understudied, which is especially critical for modalities with hierarchical semantics, e.g., video, text, etc. In this paper, we propose an Uncertainty-Adaptive Text-Video Retrieval approach, termed UATVR, which models each look-up as a distribution matching procedure. Concretely, we add additional learnable tokens in the encoders to adaptively aggregate multi-grained semantics for flexible high-level reasoning. In the refined embedding space, we represent text-video pairs as probabilistic distributions where prototypes are sampled for matching evaluation. Comprehensive experiments on four benchmarks justify the superiority of our UATVR, which achieves new state-of-the-art results on MSR-VTT (50.8%), VATEX (64.5%), MSVD (49.7%), and DiDeMo (45.8%). The code is available at https://github.com/bofang98/UATVR.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Text-to-Video Retrieval | DiDeMo | R@10.458 | 360 | |
| Text-to-Video Retrieval | MSR-VTT (test) | R@150.8 | 234 | |
| Text-to-Video Retrieval | MSVD | R@149.7 | 218 | |
| Text-to-Video Retrieval | MSVD (test) | R@146 | 204 | |
| Text-to-Video Retrieval | MSRVTT | R@147.5 | 98 | |
| Text-to-Video Retrieval | VATEX | R@164.5 | 95 | |
| Video-to-Text retrieval | MSRVTT (test) | Recall@148.1 | 15 | |
| Text-to-Video Retrieval | MSRVTT Retrieval | R@150.8 | 10 |