Research of 6‐DOF pose estimation in stacked scenes
Qi Teng, Jianjun Yi, Huazhong Li · Concurrency and Computation Practice and Experience · 2021
Summary Recently, robot vision recognition has been widely used in modern logistics system. In the face of complex and changeable environment and objects, industrial robots need to identify and operate the target objects accurately. Therefore, this article proposes a method based on Mask R‐CNN and RGB‐D to estimate the 6‐DOF of the target object. We find the 6‐DOF of the object is identified and located accurately.