DADA++: Dual Alignment Domain Adaptation for Unsupervised Video-Text Retrieval
Open Access
- 23 June 2026
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (6) , 1-23
Abstract
Video-text retrieval aims at returning the most semantically relevant videos given a textual query, which is a thriving topic in both computer vision and natural language processing communities. This article focuses on a more challenging task, i.e., Unsupervised Domain Adaptation Video-text Retrieval (UDAVR), wherein training and testing data come from different distributions. Previous approaches are mostly derived from classification-based domain adaptation methods, which are neither multi-modal nor suitable for retrieval tasks. They merely alleviate the domain shift while overlooking the pairwise misalignment issue in the target domain, i.e., there exist no semantic relationships between target videos and texts. While Foundation Models like CLIP perform well in in-domain video-text retrieval, their effectiveness significantly drops during domain shifts due to this lack of alignment. To tackle this, we propose a novel method named D ual A lignment D omain A daptation ( DADA ++). Specifically, we first introduce cross-modal semantic embedding to generate discriminative source features in a joint embedding space. Besides, we utilize cross-modal domain adaptations to balance the minimization of domain shift in a smooth manner. Furthermore, we empirically identify the pairwise misalignment in the target domain, and thus propose the i ntegrated D ual A lignment C onsistency (iDAC). The proposed iDAC adaptively aligns the video-text pairs, which are more likely to be relevant in the target domain, by verifying their cross-modal semantic proximity reciprocally in both hard and soft manners. This enables positive pairs to increase progressively while potentially aligning noisy pairs throughout the training procedure. We also provide insights into the functionality of DADA ++ through the lens of Foundation Models, explaining its superiority in a theoretical way. Compared with state-of-the-art methods, DADA ++ achieves 9.4% and 8.5% relative improvements on R@1 under the settings of TGIF \(\rightarrow\) MSR-VTT and TGIF \(\rightarrow\) MSVD, respectively, demonstrating its superior performance.
Keywords
Funding Information
- National Natural Science Foundation of China (62176081, 62106259, and 62302139)
- FRFCU China (PA2025IISL0113 and JZ2025HGTB0227)
This publication has 33 references indexed in Scilit:
- Cap4Video: What Can Auxiliary Captions Do for Text-Video Retrieval?2023
- Unsupervised person re-identification via multi-domain joint learningPattern Recognition, 2023
- X-Pool: Cross-Modal Language-Video Attention for Text-Video Retrieval2022
- Cross Modal Retrieval with Querybank Normalisation2022
- T2VLAD: Global-Local Sequence Alignment for Text-Video Retrieval2021
- Learning the Best Pooling Strategy for Visual Semantic Embedding2021
- Transferrable Prototypical Networks for Unsupervised Domain Adaptation2019
- Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation2019
- Contrastive Adaptation Network for Unsupervised Domain Adaptation2019
- Stacked Cross Attention for Image-Text Matching2018