Robust Loss Functions for Object Grasping under Limited Ground Truth
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 29 of its 45 references that are in the map. First posted on arXiv 9 Sep 2024.
Builds on
29 of 45 references.
References · 29
- 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
- 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
- 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
- 2018QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulationcited by 730
- 2017MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labelscited by 652
- 2019Symmetric Cross Entropy for Robust Learning with Noisy Labelscited by 499
- 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
- 2016Learning Hand-Eye Coordination for Robotic Grasping with Deep Learning and Large-Scale Data Collectioncited by 312
- 2018Closing the Sim-to-Real Loop: Adapting Simulation Randomization with Real World Experiencecited by 203
- 2020Rethinking Class-Balanced Methods for Long-Tailed Visual Recognition from a Domain Adaptation Perspectivecited by 143
- 2017Robotic Pick-and-Place of Novel Objects in Clutter with Multi-Affordance Grasping and Cross-Domain Image Matchingcited by 124
- 2013Deep Learning for Detecting Robotic Graspscited by 114
17 more
- 2019Grasping in the Wild:Learning 6DoF Closed-Loop Grasping from Low-Cost Demonstrationscited by 104
- 2016Sim-to-Real Robot Learning from Pixels with Progressive Netscited by 89
- 2019Curriculum Loss: Robust Learning and Generalization against Label Corruptioncited by 87
- 2019Combating Label Noise in Deep Learning Using Abstentioncited by 86
- 2018PointNetGPD: Detecting Grasp Configurations from Point Setscited by 80
- 2014Real-Time Grasp Detection Using Convolutional Neural Networkscited by 75
- 2022GraspNeRF: Multiview-based 6-DoF Grasp Detection for Transparent and Specular Objects Using Generalizable NeRFcited by 64
- 2019Antipodal Robotic Grasping using Generative Residual Convolutional Neural Networkcited by 59
- 2016Learning Dexterous Manipulation for a Soft Robotic Hand from Human Demonstrationcited by 35
- 2022Edge Grasp Network: A Graph-Based SE(3)-invariant Approach to Grasp Detectioncited by 34
- 2016High precision grasp pose detection in dense cluttercited by 33
- 2018Fully Convolutional Grasp Detection Network with Oriented Anchor Boxcited by 27
- 2016Semi-Supervised Learning with Context-Conditional Generative Adversarial Networkscited by 19
- 2022Keypoint-GraspNet: Keypoint-based 6-DoF Grasp Generation from the Monocular RGB-D inputcited by 10
- 2021Contact-Rich Manipulation of a Flexible Object based on Deep Predictive Learning using Vision and Tactilitycited by 9
- 2023DefGraspNets: Grasp Planning on 3D Fields with Graph Neural Netscited by 7
- 2023CMG-Net: An End-to-End Contact-Based Multi-Finger Dexterous Grasping Networkcited by 1
Built on by
Not cited yet.