RUI: Coherent Surface Identification and Object Recognition

点击翻译按钮获取中文摘要

The objective of the proposed work is to investigate the vision task of coherent surface identification in a single image, and to use acquired knowledge for improvement of existing algorithms. The work is a natural extension of the PI's recent work on coherent surface detection, while also adding new aspects related to the field of object recognition. The envisioned system will be able to autonomously learn object recognition models from a set of images, and then identify those objects in a complex environment. The work has strong connection to vision-based robotics as applied to unknown environments.