Zero-Shot Sketch-Based Image Retrieval with Structure-aware Asymmetric Disentanglement

TL;DR

This paper introduces STRAD, a novel structure-aware asymmetric disentanglement method for zero-shot sketch-based image retrieval, effectively handling unseen categories by disentangling structure and appearance features.

Contribution

The paper proposes a new asymmetric disentanglement approach that separates structure and appearance features, enabling effective zero-shot retrieval across unseen categories.

Findings

STRAD outperforms existing methods on three benchmark datasets.

Disentangling structure and appearance improves retrieval accuracy.

Bi-directional domain translation enhances cross-domain matching.

Abstract

The goal of Sketch-Based Image Retrieval (SBIR) is using free-hand sketches to retrieve images of the same category from a natural image gallery. However, SBIR requires all test categories to be seen during training, which cannot be guaranteed in real-world applications. So we investigate more challenging Zero-Shot SBIR (ZS-SBIR), in which test categories do not appear in the training stage. After realizing that sketches mainly contain structure information while images contain additional appearance information, we attempt to achieve structure-aware retrieval via asymmetric disentanglement.For this purpose, we propose our STRucture-aware Asymmetric Disentanglement (STRAD) method, in which image features are disentangled into structure features and appearance features while sketch features are only projected to structure space. Through disentangling structure and appearance space, bi-directional domain translation is performed between the sketch domain and the image domain. Extensive experiments demonstrate that our STRAD method remarkably outperforms state-of-the-art methods on three large-scale benchmark datasets.