Domain Disentangled Generative Adversarial Network for Zero-Shot Sketch-Based 3D Shape Retrieval

TL;DR

This paper introduces DD-GAN, a novel zero-shot sketch-based 3D shape retrieval method that disentangles domain features and synthesizes samples for unseen categories, significantly improving retrieval performance.

Contribution

The paper proposes a domain disentangled GAN that generates realistic unseen category samples using word embeddings, enabling zero-shot retrieval in sketch-based 3D shape retrieval.

Findings

Significant improvement in retrieval accuracy for unseen categories.

Effective disentanglement of domain-invariant and domain-specific features.

Enhanced discriminator discrimination using unlabeled unseen samples.

Abstract

Sketch-based 3D shape retrieval is a challenging task due to the large domain discrepancy between sketches and 3D shapes. Since existing methods are trained and evaluated on the same categories, they cannot effectively recognize the categories that have not been used during training. In this paper, we propose a novel domain disentangled generative adversarial network (DD-GAN) for zero-shot sketch-based 3D retrieval, which can retrieve the unseen categories that are not accessed during training. Specifically, we first generate domain-invariant features and domain-specific features by disentangling the learned features of sketches and 3D shapes, where the domain-invariant features are used to align with the corresponding word embeddings. Then, we develop a generative adversarial network that combines the domain-specific features of the seen categories with the aligned domain-invariant features to synthesize samples, where the synthesized samples of the unseen categories are generated by using the corresponding word embeddings. Finally, we use the synthesized samples of the unseen categories combined with the real samples of the seen categories to train the network for retrieval, so that the unseen categories can be recognized. In order to reduce the domain shift problem, we utilized unlabeled unseen samples to enhance the discrimination ability of the discriminator. With the discriminator distinguishing the generated samples from the unlabeled unseen samples, the generator can generate more realistic unseen samples. Extensive experiments on the SHREC'13 and SHREC'14 datasets show that our method significantly improves the retrieval performance of the unseen categories.