A Robotic Visual Grasping Design: Rethinking Convolution Neural Network with High-Resolutions

TL;DR

This paper introduces HRG-Net, a high-resolution, parallel-branch CNN architecture for robotic visual grasping that maintains spatial details and improves accuracy over traditional low-resolution encoding methods.

Contribution

The paper proposes a novel parallel-branch CNN design, HRG-Net, for robotic grasping that preserves high-resolution features and enhances performance compared to existing approaches.

Findings

HRG-Net improves grasping accuracy in real environments.

The parallel-branch design accelerates network training.

Maintaining high-resolution representations benefits robotic perception.

Abstract

High-resolution representations are important for vision-based robotic grasping problems. Existing works generally encode the input images into low-resolution representations via sub-networks and then recover high-resolution representations. This will lose spatial information, and errors introduced by the decoder will be more serious when multiple types of objects are considered or objects are far away from the camera. To address these issues, we revisit the design paradigm of CNN for robotic perception tasks. We demonstrate that using parallel branches as opposed to serial stacked convolutional layers will be a more powerful design for robotic visual grasping tasks. In particular, guidelines of neural network design are provided for robotic perception tasks, e.g., high-resolution representation and lightweight design, which respond to the challenges in different manipulation scenarios. We then develop a novel grasping visual architecture referred to as HRG-Net, a parallel-branch structure that always maintains a high-resolution representation and repeatedly exchanges information across resolutions. Extensive experiments validate that these two designs can effectively enhance the accuracy of visual-based grasping and accelerate network training. We show a series of comparative experiments in real physical environments at Youtube: https://youtu.be/Jhlsp-xzHFY.