Fast and Automatic Object Registration for Human-Robot Collaboration in Industrial Manufacturing

2022/04/01 by Geiß, Manuela, Baresch, Martin, Chasparis, Georgios +3
#Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2204.00597

Abstract

We present an end-to-end framework for fast retraining of object detection models in human-robot-collaboration. Our Faster R-CNN based setup covers the whole workflow of automatic image generation and labeling, model retraining on-site as well as inference on a FPGA edge device. The intervention of a human operator reduces to providing the new object together with its label and starting the training process. Moreover, we present a new loss, the intraspread-objectosphere loss, to tackle the problem of open world recognition. Though it fails to completely solve the problem, it significantly reduces the number of false positive detections of unknown objects.

Related

• Automatic Bounding Box Annotation with Small Training Data Sets for Industrial Manufacturing
• D-RMGPT: Robot-assisted collaborative tasks driven by large multimodal models
• Benchmarking Adaptive Intelligence and Computer Vision on Human-Robot Collaboration
• Robot In a Room: Toward Perfect Object Recognition in Closed Environments
• TAILOR: Teaching with Active and Incremental Learning for Object Registration