MatchNorm: Learning-based Point Cloud Registration for 6D Object Pose Estimation in the Real World
This work addresses the critical issue of deploying learning-based point cloud registration in real-world applications, such as robotics and augmented reality, by overcoming domain gaps between synthetic and real data, representing a significant advancement rather than an incremental improvement.
The paper tackles the problem of 6D object pose estimation from point clouds, where existing learning-based methods fail on real-world data due to feature distribution mismatches and sensitivity to rotation ranges; it introduces Match Normalization and a new loss function, enabling learning-based methods to achieve meaningful results on real-world datasets like TUD-L, LINEMOD, and Occluded-LINEMOD for the first time.
In this work, we tackle the task of estimating the 6D pose of an object from point cloud data. While recent learning-based approaches to addressing this task have shown great success on synthetic datasets, we have observed them to fail in the presence of real-world data. We thus analyze the causes of these failures, which we trace back to the difference between the feature distributions of the source and target point clouds, and the sensitivity of the widely-used SVD-based loss function to the range of rotation between the two point clouds. We address the first challenge by introducing a new normalization strategy, Match Normalization, and the second via the use of a loss function based on the negative log likelihood of point correspondences. Our two contributions are general and can be applied to many existing learning-based 3D object registration frameworks, which we illustrate by implementing them in two of them, DCP and IDAM. Our experiments on the real-scene TUD-L, LINEMOD and Occluded-LINEMOD datasets evidence the benefits of our strategies. They allow for the first time learning-based 3D object registration methods to achieve meaningful results on real-world data. We therefore expect them to be key to the future development of point cloud registration methods.