Sampling Using Neural Networks for colorizing the grayscale images

TL;DR

This paper explores neural network-based methods for colorizing grayscale images, comparing various generative models like CVAE, CWGAN-GP, AGE, and IVAE, and evaluates their performance on CIFAR-10 using Inception Score and human judgment.

Contribution

It introduces a comparative analysis of multiple generative models for image colorization and highlights the importance of regularization in improving sample diversity and quality.

Findings

CVAE with L1 loss and IVAE achieve highest Inception Scores

CWGAN-GP generates more diverse images but does not improve IS

Proper regularization is crucial for effective generative modeling

Abstract

The main idea of this paper is to explore the possibilities of generating samples from the neural networks, mostly focusing on the colorization of the grey-scale images. I will compare the existing methods for colorization and explore the possibilities of using new generative modeling to the task of colorization. The contributions of this paper are to compare the existing structures with similar generating structures(Decoders) and to apply the novel structures including Conditional VAE(CVAE), Conditional Wasserstein GAN with Gradient Penalty(CWGAN-GP), CWGAN-GP with L1 reconstruction loss, Adversarial Generative Encoders(AGE) and Introspective VAE(IVAE). I trained these models using CIFAR-10 images. To measure the performance, I use Inception Score(IS) which measures how distinctive each image is and how diverse overall samples are as well as human eyes for CIFAR-10 images. It turns out that CVAE with L1 reconstruction loss and IVAE achieve the highest score in IS. CWGAN-GP with L1 tends to learn faster than CWGAN-GP, but IS does not increase from CWGAN-GP. CWGAN-GP tends to generate more diverse images than other models using reconstruction loss. Also, I figured out that the proper regularization plays a vital role in generative modeling.