Continuous Surface Embeddings

TL;DR

This paper introduces a learnable, image-based dense correspondence representation that effectively maps pixels to 3D object vertices, improving dense pose estimation and scaling to various deformable objects.

Contribution

A novel, automated, learnable image-based embedding method for dense correspondences that generalizes across object categories, including animals.

Findings

Performs on par or better than state-of-the-art for human dense pose estimation.

Successfully scales to animal classes with a new dataset.

Simplifies the dense correspondence estimation process.

Abstract

In this work, we focus on the task of learning and representing dense correspondences in deformable object categories. While this problem has been considered before, solutions so far have been rather ad-hoc for specific object types (i.e., humans), often with significant manual work involved. However, scaling the geometry understanding to all objects in nature requires more automated approaches that can also express correspondences between related, but geometrically different objects. To this end, we propose a new, learnable image-based representation of dense correspondences. Our model predicts, for each pixel in a 2D image, an embedding vector of the corresponding vertex in the object mesh, therefore establishing dense correspondences between image pixels and 3D object geometry. We demonstrate that the proposed approach performs on par or better than the state-of-the-art methods for dense pose estimation for humans, while being conceptually simpler. We also collect a new in-the-wild dataset of dense correspondences for animal classes and demonstrate that our framework scales naturally to the new deformable object categories.