SwAMP: Swapped Assignment of Multi-Modal Pairs for Cross-Modal Retrieval

Abstract

We tackle the cross-modal retrieval problem, where learning is only supervised by relevant multi-modal pairs in the data. Although the contrastive learning is the most popular approach for this task, it makes potentially wrong assumption that the instances in different pairs are automatically irrelevant. To address the issue, we propose a novel loss function that is based on self-labeling of the unknown semantic classes. Specifically, we aim to predict class labels of the data instances in each modality, and assign those labels to the corresponding instances in the other modality (i.e., swapping the pseudo labels). With these swapped labels, we learn the data embedding for each modality using the supervised cross-entropy loss. This way, cross-modal instances from different pairs that are semantically related can be aligned to each other by the class predictor. We tested our approach on several real-world cross-modal retrieval problems, including text-based video retrieval, sketch-based image retrieval, and image-text retrieval. For all these tasks our method achieves significant performance improvement over the contrastive learning.

References in corpus (10)

• Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
• Microsoft COCO Captions: Data Collection and Evaluation Server
• Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models
• Momentum Contrast for Unsupervised Visual Representation Learning
• HowTo100M: Learning a Text-Video Embedding by Watching Hundred Million Narrated Video Clips
• Pixel-BERT: Aligning Image Pixels with Text by Deep Multi-Modal Transformers
• A Comprehensive Survey on Cross-modal Retrieval
• Learning a Text-Video Embedding from Incomplete and Heterogeneous Data
• Thinking Fast and Slow: Efficient Text-to-Visual Retrieval with Transformers
• Learning Language-Visual Embedding for Movie Understanding with Natural-Language