Uncertainty-Aware Cross-Modal Transfer Network for Sketch-Based 3D Shape Retrieval

TL;DR

This paper introduces UACTN, a novel network that improves sketch-based 3D shape retrieval by effectively handling noisy sketches through uncertainty-aware learning and cross-modal feature transfer.

Contribution

It proposes an uncertainty-aware transfer network that decouples sketch and shape learning, enhancing robustness to noisy data in cross-modal retrieval tasks.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Effectively handles noisy and low-quality sketch samples.

Demonstrates robustness through extensive ablation studies.

Abstract

In recent years, sketch-based 3D shape retrieval has attracted growing attention. While many previous studies have focused on cross-modal matching between hand-drawn sketches and 3D shapes, the critical issue of how to handle low-quality and noisy samples in sketch data has been largely neglected. This paper presents an uncertainty-aware cross-modal transfer network (UACTN) that addresses this issue. UACTN decouples the representation learning of sketches and 3D shapes into two separate tasks: classification-based sketch uncertainty learning and 3D shape feature transfer. We first introduce an end-to-end classification-based approach that simultaneously learns sketch features and uncertainty, allowing uncertainty to prevent overfitting noisy sketches by assigning different levels of importance to clean and noisy sketches. Then, 3D shape features are mapped into the pre-learned sketch embedding space for feature alignment. Extensive experiments and ablation studies on two benchmarks demonstrate the superiority of our proposed method compared to state-of-the-art methods.