Bilkent University
Department of Computer Engineering
CS 590/690 SEMINAR
Learning Actionable Robotic Affordances using Human Demonstrations
Dilruba Sultan Haliloğlu
Master Student
(Supervisor: Asst.Prof.Özgür Salih Öğüz)
Computer Engineering Department
Bilkent University
Abstract: Visual affordance models facilitate the perception and understanding of actionable possibilities within a robot's environment based on visual cues. However, representing the visual cues and transferring them for robotic manipulation learning is a challenging task. Our research aims to develop a methodology that integrates human demonstrations into the learning process to improve the transferability and applicability of visual affordance models in robotic manipulation tasks. We will present the latest advancement in this domain and introduce our research on learning 3D visual affordances using RGBD imagery.
DATE: March 18, Monday @ 13:50 Place: EA 502