Automatic Correction of Internal Units in Generative Neural Networks

04/13/2021
by   Ali Tousi, et al.
0

Generative Adversarial Networks (GANs) have shown satisfactory performance in synthetic image generation by devising complex network structure and adversarial training scheme. Even though GANs are able to synthesize realistic images, there exists a number of generated images with defective visual patterns which are known as artifacts. While most of the recent work tries to fix artifact generations by perturbing latent code, few investigate internal units of a generator to fix them. In this work, we devise a method that automatically identifies the internal units generating various types of artifact images. We further propose the sequential correction algorithm which adjusts the generation flow by modifying the detected artifact units to improve the quality of generation while preserving the original outline. Our method outperforms the baseline method in terms of FID-score and shows satisfactory results with human evaluation.

READ FULL TEXT

page 13

page 14

page 15

page 16

page 17

page 18

page 21

page 22

research
12/16/2021

An Unsupervised Way to Understand Artifact Generating Internal Units in Generative Neural Networks

Despite significant improvements on the image generation performance of ...
research
01/17/2022

Can We Find Neurons that Cause Unrealistic Images in Deep Generative Networks?

Even though image generation with Generative Adversarial Networks has be...
research
12/21/2018

Non-Adversarial Image Synthesis with Generative Latent Nearest Neighbors

Unconditional image generation has recently been dominated by generative...
research
12/12/2019

An Efficient Explorative Sampling Considering the Generative Boundaries of Deep Generative Neural Networks

Deep generative neural networks (DGNNs) have achieved realistic and high...
research
09/29/2020

TinyGAN: Distilling BigGAN for Conditional Image Generation

Generative Adversarial Networks (GANs) have become a powerful approach f...
research
08/20/2019

Sparse Generative Adversarial Network

We propose a new approach to Generative Adversarial Networks (GANs) to a...
research
06/17/2022

SOS: Score-based Oversampling for Tabular Data

Score-based generative models (SGMs) are a recent breakthrough in genera...

Please sign up or login with your details

Forgot password? Click here to reset