A Controllable Appearance Representation for Flexible Transfer and Editing
This work addresses the need for intuitive and controllable appearance manipulation in computer graphics and vision, offering a domain-specific solution for material editing and transfer.
The paper tackles the problem of material appearance representation and editing by developing a self-supervised, interpretable latent space using an adapted FactorVAE, which enables flexible transfer and editing of appearance attributes like hue or glossiness onto target geometries with fine-grained control.
We present a method that computes an interpretable representation of material appearance within a highly compact, disentangled latent space. This representation is learned in a self-supervised fashion using an adapted FactorVAE. We train our model with a carefully designed unlabeled dataset, avoiding possible biases induced by human-generated labels. Our model demonstrates strong disentanglement and interpretability by effectively encoding material appearance and illumination, despite the absence of explicit supervision. Then, we use our representation as guidance for training a lightweight IP-Adapter to condition a diffusion pipeline that transfers the appearance of one or more images onto a target geometry, and allows the user to further edit the resulting appearance. Our approach offers fine-grained control over the generated results: thanks to the well-structured compact latent space, users can intuitively manipulate attributes such as hue or glossiness in image space to achieve the desired final appearance.