CVJun 14, 2020

2D Image Relighting with Image-to-Image Translation

arXiv:2006.07816v214 citations
Originality Synthesis-oriented
AI Analysis

This addresses image manipulation for computer vision applications, but is incremental as it builds on existing GAN methods and datasets.

The paper tackled the ill-posed problem of 2D image relighting by changing light source direction using GANs, specifically pix2pix trained on the VIDIT dataset, resulting in 8 neural networks capable of converting any direction to one of 8 specific ones and a CNN for light direction identification.

With the advent of Generative Adversarial Networks (GANs), a finer level of control in manipulating various features of an image has become possible. One example of such fine manipulation is changing the position of the light source in a scene. This is fundamentally an ill-posed problem, since it requires understanding the scene geometry to generate proper lighting effects. This problem is not a trivial one and can become even more complicated if we want to change the direction of the light source from any direction to a specific one. Here we provide our attempt to solve this problem using GANs. Specifically, pix2pix [arXiv:1611.07004] trained with the dataset VIDIT [arXiv:2005.05460] which contains images of the same scene with different types of light temperature and 8 different light source positions (N, NE, E, SE, S, SW, W, NW). The results are 8 neural networks trained to be able to change the direction of the light source from any direction to one of the 8 previously mentioned. Additionally, we provide, as a tool, a simple CNN trained to identify the direction of the light source in an image.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes