Uncertainty-Aware Cross-Modal Transfer Network for Sketch-Based 3D Shape Retrieval
This work addresses a critical issue in cross-modal retrieval for computer vision applications, but it is incremental as it builds on existing methods by adding uncertainty handling.
The paper tackles the problem of handling low-quality and noisy sketches in sketch-based 3D shape retrieval by proposing an uncertainty-aware cross-modal transfer network (UACTN), which learns sketch features with uncertainty to prevent overfitting and achieves superior performance compared to state-of-the-art methods on two benchmarks.
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.