Skip to ContentSkip to Navigation
How to find us S.H. (Hamidreza) Mohades Kasaei, PhD

Research interests

My research interest lies at the intersection of machine learning, robotics, and machine vision, particularly, at the area of open-ended learning, 3D object perception, grasp affordance detection, and object manipulation. I am interested in developing algorithms for an adaptive perception system based on interactive environment exploration and open-ended learning, which enables robots to learn from past experiences and interact with human users. I have been investigating on active perception, where robots use their mobility and manipulation capabilities not only to gain perceptual information to model the world but also to predict the next-best-view for improving object detection and manipulation performances. I have evaluated my works on different robotic platforms, including PR2, robotic arms (UR5e, Kinova, Franka), and humanoid robots. For more details, please check my webpage  www.ai.rug.nl/hkasaei

Publications

Local-HDP: Interactive Open-ended 3D Object Category Recognition in Real-Time Robotic Scenarios

Self-Imitation Learning by Planning

Accelerating Reinforcement Learning for Reaching using Continuous Curriculum Learning

Investigating the Importance of Shape Features, Color Constancy, Color Spaces and Similarity Measures in Open-Ended 3D Object Recognition

Learning to Grasp 3D Objects using Deep Residual U-Nets

Local-LDA: Open-Ended Learning of Latent Topics for 3D Object Recognition

The State of Service Robots: Current Bottlenecks in Object Perception and Manipulation

Interactive Open-Ended Object, Affordance and Grasp Learning for Robotic Manipulation

Look Further to Recognize Better: Learning Shared Topics and Category-Specific Dictionaries for Open-Ended 3D Object Recognition

OrthographicNet: A deep transfer learning approach for 3D object recognition in open-ended domains

Read more