Introducing Pose Consistency and Warp-Alignment for Self-Supervised 6D Object Pose Estimation in Color Images

TL;DR

This paper introduces a self-supervised framework for 6D object pose estimation that enhances generalization from synthetic data without requiring real-world pose annotations, using pose and photometric consistency techniques.

Contribution

It proposes a two-stage self-supervised approach that improves 6D pose estimation by enforcing pose and photometric consistency, applicable on top of existing neural networks without real image annotations.

Findings

Achieves state-of-the-art results on LINEMOD, LINEMOD OCCLUSION, and HomebrewedDB datasets.

Outperforms methods trained only on synthetic data and domain adaptation baselines.

Effective in real-world scenarios without requiring pose annotations or depth information.

Abstract

Most successful approaches to estimate the 6D pose of an object typically train a neural network by supervising the learning with annotated poses in real world images. These annotations are generally expensive to obtain and a common workaround is to generate and train on synthetic scenes, with the drawback of limited generalisation when the model is deployed in the real world. In this work, a two-stage 6D object pose estimator framework that can be applied on top of existing neural-network-based approaches and that does not require pose annotations on real images is proposed. The first self-supervised stage enforces the pose consistency between rendered predictions and real input images, narrowing the gap between the two domains. The second stage fine-tunes the previously trained model by enforcing the photometric consistency between pairs of different object views, where one image is warped and aligned to match the view of the other and thus enabling their comparison. In the absence of both real image annotations and depth information, applying the proposed framework on top of two recent approaches results in state-of-the-art performance when compared to methods trained only on synthetic data, domain adaptation baselines and a concurrent self-supervised approach on LINEMOD, LINEMOD OCCLUSION and HomebrewedDB datasets.