Feature Quantization Improves GAN Training

04/05/2020
by   fcq, et al.
10

The instability in GAN training has been a long-standing problem despite remarkable research efforts. We identify that instability issues stem from difficulties of performing feature matching with mini-batch statistics, due to a fragile balance between the fixed target distribution and the progressively generated distribution. In this work, we propose Feature Quantization (FQ) for the discriminator, to embed both true and fake data samples into a shared discrete space. The quantized values of FQ are constructed as an evolving dictionary, which is consistent with feature statistics of the recent distribution history. Hence, FQ implicitly enables robust feature matching in a compact space. Our method can be easily plugged into existing GAN models, with little computational overhead in training. We apply FQ to 3 representative GAN models on 9 benchmarks: BigGAN for image generation, StyleGAN for face synthesis, and U-GAT-IT for unsupervised image-to-image translation. Extensive experimental results show that the proposed FQ-GAN can improve the FID scores of baseline methods by a large margin on a variety of tasks, achieving new state-of-the-art performance.

READ FULL TEXT

page 8

page 14

page 15

page 16

page 17

page 18

page 19

page 20

research
03/26/2020

Image Generation Via Minimizing Fréchet Distance in Discriminator Feature Space

For a given image generation problem, the intrinsic image manifold is of...
research
06/18/2020

Progressively Unfreezing Perceptual GAN

Generative adversarial networks (GANs) are widely used in image generati...
research
05/30/2023

Towards Accurate Data-free Quantization for Diffusion Models

In this paper, we propose an accurate data-free post-training quantizati...
research
02/29/2020

Reusing Discriminators for Encoding Towards Unsupervised Image-to-Image Translation

Unsupervised image-to-image translation is a central task in computer vi...
research
02/25/2019

Harmonizing Maximum Likelihood with GANs for Multimodal Conditional Generation

Recent advances in conditional image generation tasks, such as image-to-...
research
08/23/2021

Exploring the Quality of GAN Generated Images for Person Re-Identification

Recently, GAN based method has demonstrated strong effectiveness in gene...
research
07/26/2021

A Unified Hyper-GAN Model for Unpaired Multi-contrast MR Image Translation

Cross-contrast image translation is an important task for completing mis...

Please sign up or login with your details

Forgot password? Click here to reset