Manifold Learning Benefits GANs

2021-12-23 · CVPR 2022 1 · Yao Ni, Piotr Koniusz, Richard Hartley, Richard Nock

In this paper, we improve Generative Adversarial Networks by incorporating a manifold learning step into the discriminator. We consider locality-constrained linear and subspace-based manifolds, and locality-constrained non-linear manifolds. In our design, the manifold learning and coding steps are intertwined with layers of the discriminator, with the goal of attracting intermediate feature representations onto manifolds. We adaptively balance the discrepancy between feature representations and their manifold view, which is a trade-off between denoising on the manifold and refining the manifold. We find that locality-constrained non-linear manifolds outperform linear manifolds due to their non-uniform density and smoothness. We also substantially outperform state-of-the-art baselines.

📄 PDF Abstract BibTeX arXiv:2112.12618

Code (0)

등록된 구현이 없습니다.

Tasks

Denoising

Manifold-preserved GANs

2021-09-18 · Haozhe Liu, Hanbang Liang, Xianxu Hou, Haoqian Wu 외

Generative Adversarial Networks (GANs) have been widely adopted in various fields. However, existing GANs generally are not able to preserve the manifold of data space, mainly due to the simple representation of discrimi…

Intriguing Properties of Modern GANs

2024-02-21 · Roy Friedman, Yair Weiss

Modern GANs achieve remarkable performance in terms of generating realistic and diverse samples. This has led many to believe that ``GANs capture the training data manifold''. In this work we show that this interpretatio…

MR-GAN: Manifold Regularized Generative Adversarial Networks

2018-11-22 · Qunwei Li, Bhavya Kailkhura, Rushil Anirudh, Yi Zhou 외

Despite the growing interest in generative adversarial networks (GANs), training GANs remains a challenging problem, both from a theoretical and a practical standpoint. To address this challenge, in this paper, we propos…

SMaRt: Improving GANs with Score Matching Regularity

2023-11-30 · Mengfei Xia, Yujun Shen, Ceyuan Yang, Ran Yi 외

Generative adversarial networks (GANs) usually struggle in learning from highly diverse data, whose underlying manifold is complex. In this work, we revisit the mathematical foundations of GANs, and theoretically reveal …

valid

Disconnected Manifold Learning for Generative Adversarial Networks

2018-06-03 · NeurIPS 2018 12 · Mahyar Khayatkhoei, Ahmed Elgammal, Maneesh Singh

Natural images may lie on a union of disjoint manifolds rather than one globally connected manifold, and this can cause several difficulties for the training of common Generative Adversarial Networks (GANs). In this work…