Unsupervised Landmark Discovery Using Consistency Guided Bottleneck

TL;DR

This paper introduces a novel consistency-guided bottleneck approach for unsupervised object landmark discovery, leveraging pseudo-supervision and adaptive heatmaps to improve robustness and accuracy across diverse datasets.

Contribution

It proposes a new consistency-guided bottleneck mechanism that uses landmark correspondence and pseudo-supervision to generate adaptive, structurally informed heatmaps in an unsupervised setting.

Findings

Outperforms existing methods on five diverse datasets.

Effectively leverages landmark consistency for robustness.

Provides state-of-the-art results in unsupervised landmark discovery.

Abstract

We study a challenging problem of unsupervised discovery of object landmarks. Many recent methods rely on bottlenecks to generate 2D Gaussian heatmaps however, these are limited in generating informed heatmaps while training, presumably due to the lack of effective structural cues. Also, it is assumed that all predicted landmarks are semantically relevant despite having no ground truth supervision. In the current work, we introduce a consistency-guided bottleneck in an image reconstruction-based pipeline that leverages landmark consistency, a measure of compatibility score with the pseudo-ground truth to generate adaptive heatmaps. We propose obtaining pseudo-supervision via forming landmark correspondence across images. The consistency then modulates the uncertainty of the discovered landmarks in the generation of adaptive heatmaps which rank consistent landmarks above their noisy counterparts, providing effective structural information for improved robustness. Evaluations on five diverse datasets including MAFL, AFLW, LS3D, Cats, and Shoes demonstrate excellent performance of the proposed approach compared to the existing state-of-the-art methods. Our code is publicly available at https://github.com/MamonaAwan/CGB_ULD.