On Object Symmetries and 6D Pose Estimation from Images
This addresses a challenge in computer vision for robotics and industrial applications, but it is incremental as it builds on existing pose estimation frameworks.
The paper tackles the problem of 6D pose estimation from images for symmetrical objects, which are often overlooked, by proposing a normalization of pose rotation that works with any algorithm. It validates the method on synthetic and real datasets from T-Less, showing benefits for symmetrical and almost symmetrical objects.
Objects with symmetries are common in our daily life and in industrial contexts, but are often ignored in the recent literature on 6D pose estimation from images. In this paper, we study in an analytical way the link between the symmetries of a 3D object and its appearance in images. We explain why symmetrical objects can be a challenge when training machine learning algorithms that aim at estimating their 6D pose from images. We propose an efficient and simple solution that relies on the normalization of the pose rotation. Our approach is general and can be used with any 6D pose estimation algorithm. Moreover, our method is also beneficial for objects that are 'almost symmetrical', i.e. objects for which only a detail breaks the symmetry. We validate our approach within a Faster-RCNN framework on a synthetic dataset made with objects from the T-Less dataset, which exhibit various types of symmetries, as well as real sequences from T-Less.