Freeview Sketching: View-Aware Fine-Grained Sketch-Based Image Retrieval

TL;DR

This paper introduces a view-aware approach for fine-grained sketch-based image retrieval, addressing viewpoint variability by leveraging multi-view 3D projections and disentangled features to improve retrieval accuracy.

Contribution

It proposes a novel view-aware system that combines multi-view 3D projections and disentangled features for flexible, view-specific and view-agnostic sketch retrieval.

Findings

Enhanced retrieval accuracy on standard datasets.

Effective handling of viewpoint variations in sketches.

User preference for view-specific retrieval is accommodated.

Abstract

In this paper, we delve into the intricate dynamics of Fine-Grained Sketch-Based Image Retrieval (FG-SBIR) by addressing a critical yet overlooked aspect -- the choice of viewpoint during sketch creation. Unlike photo systems that seamlessly handle diverse views through extensive datasets, sketch systems, with limited data collected from fixed perspectives, face challenges. Our pilot study, employing a pre-trained FG-SBIR model, highlights the system's struggle when query-sketches differ in viewpoint from target instances. Interestingly, a questionnaire however shows users desire autonomy, with a significant percentage favouring view-specific retrieval. To reconcile this, we advocate for a view-aware system, seamlessly accommodating both view-agnostic and view-specific tasks. Overcoming dataset limitations, our first contribution leverages multi-view 2D projections of 3D objects, instilling cross-modal view awareness. The second contribution introduces a customisable cross-modal feature through disentanglement, allowing effortless mode switching. Extensive experiments on standard datasets validate the effectiveness of our method.