Revisiting Fully Convolutional Geometric Features for Object 6D Pose Estimation

TL;DR

This paper enhances 6D object pose estimation by revisiting Fully Convolutional Geometric Features, focusing on point-level discriminative features, leading to state-of-the-art results through tailored loss functions and data augmentations.

Contribution

It introduces modifications to FCGF for improved 6D pose estimation, emphasizing point-level features and optimized training strategies.

Findings

Outperforms recent methods on benchmark datasets

Effective use of contrastive loss and data augmentation

Thorough ablation studies validate each modification

Abstract

Recent works on 6D object pose estimation focus on learning keypoint correspondences between images and object models, and then determine the object pose through RANSAC-based algorithms or by directly regressing the pose with end-to-end optimisations. We argue that learning point-level discriminative features is overlooked in the literature. To this end, we revisit Fully Convolutional Geometric Features (FCGF) and tailor it for object 6D pose estimation to achieve state-of-the-art performance. FCGF employs sparse convolutions and learns point-level features using a fully-convolutional network by optimising a hardest contrastive loss. We can outperform recent competitors on popular benchmarks by adopting key modifications to the loss and to the input data representations, by carefully tuning the training strategies, and by employing data augmentations suitable for the underlying problem. We carry out a thorough ablation to study the contribution of each modification. The code is available at https://github.com/jcorsetti/FCGF6D.