Transformation Consistency Regularization- A Semi-Supervised Paradigm for Image-to-Image Translation

TL;DR

This paper introduces a semi-supervised learning approach called Transformation Consistency Regularization for image-to-image translation tasks, leveraging geometric transformations to improve data efficiency and performance with limited labeled data.

Contribution

It extends consistency regularization to image-to-image translation, a challenging setting, using diverse geometric transformations to enforce prediction invariance.

Findings

Achieves similar results to fully-supervised methods with only 10-20% labeled data.

Effective in image colorization, denoising, and super-resolution tasks.

Enhances video processing by leveraging few labeled frames.

Abstract

Scarcity of labeled data has motivated the development of semi-supervised learning methods, which learn from large portions of unlabeled data alongside a few labeled samples. Consistency Regularization between model's predictions under different input perturbations, particularly has shown to provide state-of-the art results in a semi-supervised framework. However, most of these method have been limited to classification and segmentation applications. We propose Transformation Consistency Regularization, which delves into a more challenging setting of image-to-image translation, which remains unexplored by semi-supervised algorithms. The method introduces a diverse set of geometric transformations and enforces the model's predictions for unlabeled data to be invariant to those transformations. We evaluate the efficacy of our algorithm on three different applications: image colorization, denoising and super-resolution. Our method is significantly data efficient, requiring only around 10 - 20% of labeled samples to achieve similar image reconstructions to its fully-supervised counterpart. Furthermore, we show the effectiveness of our method in video processing applications, where knowledge from a few frames can be leveraged to enhance the quality of the rest of the movie.