Unsupervised Learning of Object Landmarks through Conditional Image Generation

TL;DR

This paper introduces an unsupervised method for learning object landmarks by generating images conditioned on appearance and geometry, effectively capturing key features without manual labels across diverse datasets.

Contribution

The authors present a novel unsupervised approach that learns object landmarks through conditional image generation, outperforming existing methods and applicable to various object types.

Findings

Successfully learned landmarks from synthetic deformations and videos

Outperformed state-of-the-art unsupervised landmark detectors

Applicable to diverse datasets including faces, objects, and digits

Abstract

We propose a method for learning landmark detectors for visual objects (such as the eyes and the nose in a face) without any manual supervision. We cast this as the problem of generating images that combine the appearance of the object as seen in a first example image with the geometry of the object as seen in a second example image, where the two examples differ by a viewpoint change and/or an object deformation. In order to factorize appearance and geometry, we introduce a tight bottleneck in the geometry-extraction process that selects and distils geometry-related features. Compared to standard image generation problems, which often use generative adversarial networks, our generation task is conditioned on both appearance and geometry and thus is significantly less ambiguous, to the point that adopting a simple perceptual loss formulation is sufficient. We demonstrate that our approach can learn object landmarks from synthetic image deformations or videos, all without manual supervision, while outperforming state-of-the-art unsupervised landmark detectors. We further show that our method is applicable to a large variety of datasets - faces, people, 3D objects, and digits - without any modifications.