Benchmarking Convolutional Neural Networks for Object Segmentation and Pose Estimation

Convolutional neural networks (CNNs), particularly those designed for object segmentation and pose estimation, are now applied to robotics applications involving mobile manipulation. For these robotic applications to be successful, robust and accurate performance from the CNNs is critical. Therefore, in order to develop an understanding of CNN performance, several CNN architectures are benchmarked on a set of metrics for object segmentation and pose estimation. This paper presents these benchmarking results, which show that metric performance is dependent on the complexity of network architectures. These findings can be used to guide and improve the development of CNNs for object segmentation and pose estimation in the future.