3D Highlighter Accepted to CVPR 2023 ⭐

3D Highlighter reasons about where to place seemingly unrelated concepts in semantically meaningful locations on the 3D shape, such as a 'poncho' on a horse or 'shoes' on an alien. Our method contextualizes the text description using a neural field and colors the corresponding region of the shape using a probability-weighted blend.

DA Wand Accepted to CVPR 2023 ⭐

DA Wand is a framework which enables training a neural network to learn a data-driven prior for distortion-aware mesh segmentation. Key to our method is a probability-guided mesh parameterization layer, which is differentiable, fast, and theoretically grounded.