D3still: Decoupled Differential Distillation for Asymmetric Image Retrieval

Existing methods for asymmetric image retrieval employ a rigid pairwise similarity constraint between the query network and the larger gallery network. However, these one-to-one constraint approaches often fail to maintain retrieval order consistency, especially when the query network has limited representational capacity. To overcome this problem, we introduce the Decoupled Differential Distillation (D3still) framework. This framework shifts from absolute one-to-one supervision to optimizing the relational differences in pairwise similarities produced by the query and gallery networks, thereby preserving a consistent retrieval order across both networks. Our method involves computing a pairwise similarity differential matrix within the gallery domain, which is then decomposed into three components: feature representation knowledge, inconsistent pairwise similarity differential knowledge, and consistent pairwise similarity differential knowledge. This strategic decomposition aligns the retrieval ranking of the query network with the gallery network effectively. Extensive experiments on various bench-mark datasets reveal that D3still surpasses state-of-the-art methods in asymmetric image retrieval. Code is available at https://github.com/SCY-X/D3still.

1. Introduction

The predominant image retrieval methods [1], [32], [34] based on deep learning, typically involve mapping both query and gallery images into a shared feature space that is highly discriminative. Within this space, gallery images are then ranked according to their relevance to the query image. How-ever, this feature mapping process often relies on large neural networks, which pose practical challenges for deployment on edge devices in real-world scenarios. Consequently, this ne-cessitates uploading query images to cloud-based platforms for feature extraction, resulting in dependencies on network connectivity and additional computational overhead.