InverseRenderNet: Learning single image inverse rendering

Understand

We show how to train a fully convolutional neural network to perform inverse rendering from a single, uncontrolled image.

• The network takes an RGB image as input, regresses albedo and normal maps from which we compute lighting coefficients.
• Our network is trained using large uncontrolled image collections without ground truth.
• By incorporating a differentiable renderer, our network can learn from self-supervision.