Dynamic Visual Semantic Sub-Embeddings and Fast Re-Ranking for Image-Text Retrieval

The core of image-text retrieval is to accurately measure the similarity between different modalities in a unified representation space. However, compared to textual descriptions of a certain perspective, the visual modality has more semantic variations. Therefore, images are usually associated with multiple textual captions in databases. Although popular symmetric embedding methods have explored numerous modal interaction approaches, they often learn toward outputting the average representation of multiple semantic variations within image embeddings. Consequently, information entropy in embeddings is increased, resulting in redundancy and decreased accuracy. In this work, we propose a Dynamic Visual Semantic Sub-Embeddings framework (DVSE) to reduce the information entropy. Specifically, we obtain a set of heterogeneous visual sub-embeddings through dynamic orthogonal constraint loss. To encourage the generated candidate image embeddings to capture various semantic variations, we construct a mixed distribution and employ a variance-aware weighting loss to assign different weights to the optimization process. In addition, we develop a Fast Re-ranking strategy (FR) to efficiently evaluate the retrieval results and enhance the performance. We compare the performance with existing set-based method using five image feature encoders and three text feature encoders on three benchmark datasets: MSCOCO, Flickr30K and CUB Captions. We also show the role of different components by ablation studies and perform a sensitivity analysis of the hyperparameters. The qualitative analysis of visualized bidirectional retrieval and attention maps further demonstrates the ability of our method to encode semantic variations.