HAL: Improved Text-Image Matching by Mitigating Visual Semantic Hubs

Abstract

The hubness problem widely exists in high-dimensional embedding space and is a fundamental source of error for crossmodal matching tasks. In this work, we study the emergence of hubs in Visual Semantic Embeddings (VSE) with application to text-image matching. We analyze the pros and cons of two widely adopted optimization objectives for training VSE and propose a novel hubness-aware loss function (HAL) that addresses previous methods' defects. Unlike (Faghri et al. 2018) which simply takes the hardest sample within a minibatch, HAL takes all samples into account, using both local and global statistics to scale up the weights of "hubs". We experiment our method with various configurations of model architectures and datasets. The method exhibits exceptionally good robustness and brings consistent improvement on the task of text-image matching across all settings. Specifically, under the same model architectures as (Faghri et al. 2018) and(Lee et al. 2018), by switching only the learning objective, we report a maximum R@1 improvement of 7.4% on MS-COCO and 8.3% on Flickr30k. 1

Adaptive Offline Quintuplet Loss for Image-Text Matching

“…A small white clock sitting on top of a wooden table effectively cooperate with an online triplet loss, leading to significant improvement. It should be noticed that Liu et al [19] explicitly feed adaptive penalty weight into triplet loss for image-text matching. However, they use it to solve the hubness problem, while we implicitly feed hierarchical information into the model to enlarge the similarity score differences among different pair classes.…”