Semantically Tied Paired Cycle Consistency for Any-Shot Sketch-based Image Retrieval
This addresses the problem of retrieving images from sketches with limited training data for applications in computer vision, though it is incremental as it builds on prior GAN and cycle consistency methods.
The paper tackles any-shot sketch-based image retrieval by proposing a semantically aligned paired cycle-consistent GAN that maps sketches and images to a common semantic space without needing aligned pairs, achieving significant performance boosts over state-of-the-art on extended Sketchy, TU-Berlin, and QuickDraw datasets.
Low-shot sketch-based image retrieval is an emerging task in computer vision, allowing to retrieve natural images relevant to hand-drawn sketch queries that are rarely seen during the training phase. Related prior works either require aligned sketch-image pairs that are costly to obtain or inefficient memory fusion layer for mapping the visual information to a semantic space. In this paper, we address any-shot, i.e. zero-shot and few-shot, sketch-based image retrieval (SBIR) tasks, where we introduce the few-shot setting for SBIR. For solving these tasks, we propose a semantically aligned paired cycle-consistent generative adversarial network (SEM-PCYC) for any-shot SBIR, where each branch of the generative adversarial network maps the visual information from sketch and image to a common semantic space via adversarial training. Each of these branches maintains cycle consistency that only requires supervision at the category level, and avoids the need of aligned sketch-image pairs. A classification criteria on the generators' outputs ensures the visual to semantic space mapping to be class-specific. Furthermore, we propose to combine textual and hierarchical side information via an auto-encoder that selects discriminating side information within a same end-to-end model. Our results demonstrate a significant boost in any-shot SBIR performance over the state-of-the-art on the extended version of the challenging Sketchy, TU-Berlin and QuickDraw datasets.