FS6D: Few-Shot 6D Pose Estimation of Novel Objects
This addresses the scalability issue in 6D pose estimation for robotics and AR/VR by enabling pose estimation of novel objects without extensive training, though it is incremental in improving generalization.
The paper tackles the problem of few-shot 6D pose estimation for unknown objects using only a few support views, proposing a dense prototypes matching framework with transformers and a large-scale RGBD dataset for pre-training, achieving competitive results on benchmarks.
6D object pose estimation networks are limited in their capability to scale to large numbers of object instances due to the close-set assumption and their reliance on high-fidelity object CAD models. In this work, we study a new open set problem; the few-shot 6D object poses estimation: estimating the 6D pose of an unknown object by a few support views without extra training. To tackle the problem, we point out the importance of fully exploring the appearance and geometric relationship between the given support views and query scene patches and propose a dense prototypes matching framework by extracting and matching dense RGBD prototypes with transformers. Moreover, we show that the priors from diverse appearances and shapes are crucial to the generalization capability under the problem setting and thus propose a large-scale RGBD photorealistic dataset (ShapeNet6D) for network pre-training. A simple and effective online texture blending approach is also introduced to eliminate the domain gap from the synthesis dataset, which enriches appearance diversity at a low cost. Finally, we discuss possible solutions to this problem and establish benchmarks on popular datasets to facilitate future research. The project page is at \url{https://fs6d.github.io/}.