This work proposes first a perception process for 6D pose estimation, which returns a discrete distribution of object poses in a scene.
Tracking the 6D pose of objects in video sequences is important for robot manipulation.
Ranked #2 on 6D Pose Estimation on YCB-Video
The effectiveness of the proposed approach is demonstrated by developing a robotic system that picks a previously unseen object from a table-top and places it in a constrained space.
The hand's point cloud is pruned and robust global registration is performed to generate object pose hypotheses, which are clustered.
This work proposes a set of interpretive principles for how a robotic arm can use pointing actions to communicate task information to people by extending existing models from the related literature.
To evaluate this method, a dataset of densely packed objects with challenging setups for state-of-the-art approaches is collected.
This work proposes an autonomous process for pose estimation that spans from data generation to scene-level reasoning and self-learning.
We then show that the performance of the detector can be substantially improved by using a small set of weakly annotated real images, where a human provides only a list of objects present in each image without indicating the location of the objects.
The pointsets are then matched to congruent sets on the 3D object model to generate pose estimates.
Experimental results indicate that this process is able to quickly identify in cluttered scenes physically-consistent object poses that are significantly closer to ground truth compared to poses found by point cloud registration methods.
The models are placed in physically realistic poses with respect to their environment to generate a labeled synthetic dataset.