Robust Loss Functions for Object Grasping under Limited Ground Truth

TL;DR

This paper introduces robust loss functions for object grasping neural networks that effectively handle missing and noisy ground truth data, significantly improving accuracy in practical robotic applications.

Contribution

It proposes a novel predicted category probability method for unlabeled data and a symmetric loss function to resist label noise, enhancing neural network robustness.

Findings

Performance improved by 2-13% with the new loss functions.

The methods are effective with typical grasping neural networks.

Loss functions are easy to implement and robust against data imperfections.

Abstract

Object grasping is a crucial technology enabling robots to perceive and interact with the environment sufficiently. However, in practical applications, researchers are faced with missing or noisy ground truth while training the convolutional neural network, which decreases the accuracy of the model. Therefore, different loss functions are proposed to deal with these problems to improve the accuracy of the neural network. For missing ground truth, a new predicted category probability method is defined for unlabeled samples, which works effectively in conjunction with the pseudo-labeling method. Furthermore, for noisy ground truth, a symmetric loss function is introduced to resist the corruption of label noises. The proposed loss functions are powerful, robust, and easy to use. Experimental results based on the typical grasping neural network show that our method can improve performance by 2 to 13 percent.