Michael Villamizar

About

Michael Villamizar is a robotics and computer vision researcher whose work sits at the intersection of human-robot interaction (HRI), machine learning, and autonomous systems. His research has made significant contributions to how robots perceive, learn from, and interact with humans in real-world environments. Villamizar is perhaps best known for pioneering work on proactive robot behavior and human-assisted learning, with foundational papers from 2013 and 2017 accumulating nearly 70 citations combined. His key insight — that robots can improve their capabilities through structured, minimally intrusive human guidance — has shaped how researchers approach interactive machine learning. His Online Random Ferns classifier exemplifies this philosophy, enabling robots to build robust object models with limited human supervision. In computer vision, Villamizar has advanced depth-based human pose estimation using Convolutional Neural Networks, producing fast, reliable multi-person pose detection suited for HRI scenarios, with his 2018 paper earning 32 citations. His 2017 Aerial Social Force Model extended social robotics into three dimensions, enabling autonomous drones to safely accompany humans in urban settings. Spanning over a decade of research, Villamizar's body of work — totaling nearly 200 citations — reflects a consistent commitment to making robots more socially intelligent, perceptually capable, and genuinely useful alongside people.

Research Focus

Key Achievements

Top Papers

1. 1
   Teaching Robot’s Proactive Behavior Using Human Assistance

   42 citations · 2017
2. 2
   Real-time Convolutional Networks for Depth-based Human Pose Estimation

   32 citations · 2018
3. 3
   Proactive behavior of an autonomous mobile robot for human-assisted learning

   26 citations · 2013
4. 4
   Aerial social force model: A new framework to accompany people using autonomous flying robots

   23 citations · 2017
5. 5
   Online Human-AssistedLearning using Random Ferns

   18 citations · 2013
6. 6
   Combining color-based invariant gradient detector with HoG descriptors for robust image detection in scenes under cast shadows

   18 citations · 2009
7. 7
   Robot Interactive Learning through Human Assistance

   14 citations · 2013
8. 8
   Interactive multiple object learning with scanty human supervision

   10 citations · 2016
9. 9
   Fast online learning and detection of natural landmarks for autonomous aerial robots

   9 citations · 2014
10. 10
    Residual Pose: A Decoupled Approach for Depth-based 3D Human Pose Estimation

    4 citations · 2020

Key Collaborators

Contact & Links