Pablo Azagra

About

Pablo Azagra is a robotics researcher whose work sits at the intersection of human-robot interaction, incremental learning, and multimodal perception. His primary research focuses on enabling robots to learn new object models naturally during interactions with humans, without requiring pre-programmed knowledge. Azagra’s most influential contribution is an end-to-end pipeline for incremental object learning from natural human-robot interactions, detailed in his top-cited paper (17 citations). He has also developed several multimodal datasets—including one with synchronized stereo audio and multi-camera recordings—that are essential for training robots to recognize objects, segment regions of interest, and understand interaction types in real time. More recently, Azagra has extended his expertise to person re-identification, proposing a self-adaptive gallery construction method for open-world scenarios, a key capability for robotic tracking and navigation tasks. With over 38 total citations across his most cited works, Azagra is building foundational tools for robots that learn continuously and autonomously in unstructured, human-centered environments.

Research Focus

Key Achievements

Top Papers

1. 1
   Incremental Learning of Object Models From Natural Human–Robot Interactions

   17 citations · 2020
2. 2
   A multimodal dataset for object model learning from natural human-robot interaction

   11 citations · 2017
3. 3
   A Self-Adaptive Gallery Construction Method for Open-World Person Re-Identification

   4 citations · 2023
4. 4
   Finding Regions of Interest from Multimodal Human-Robot Interactions

   2 citations · 2017
5. 5
   A Multimodal Human-Robot Interaction Dataset

   2 citations · 2016
6. 6
   A Multimodal Dataset for Interactive and Incremental Learning of Object Models

   2 citations · 2016

Key Collaborators

Contact & Links