Juanpablo Heredia

Ph.D. Candidate

prof_pic.png

I am a PhD student in data science, currently affiliated with the Visual Data Science Lab at Fundação Getulio Vargas (FGV), Rio de Janeiro, Brazil, under the supervision of Prof. Jorge Poco. I hold a full doctoral scholarship from FAPESP (São Paulo Research Foundation) in the Direct Doctoral modality.

My work focuses on spatiotemporal data analysis and visual analytics, with particular interest in geometric deep learning and explainable AI.

Previously, I was a researcher in the RUTAS project at Universidad Católica San Pablo (UCSP) in Arequipa, Peru, where I worked on computer vision and multimodal deep learning. I hold a BSc in Computer Science (2021) from the Computer Science Department at UCSP.

selected publications

  1. topoxignalsposter.png
    Topology-Aware Visual Analysis of GNN Explanations in Dynamic Graphs
    Juanpablo Heredia and Jorge Poco
    Computer Graphics Forum (Proc. EuroVis) - Posters and Demos, 2026
  2. urbanclipatlas.png
    URBANCLIPATLAS: A Visual Analytics Framework for Event and Scene Retrieval in Urban Videos
    Joel Perca, Luis Sante Taipe, Juanpablo Heredia, and 3 more authors
    Computer Graphics Forum (Proc. EuroVis), 2026
  3. xallm.png
    Interactive Exploration and Explanation of Spatio-Temporal Anomalies with Graph-LLM Integration
    Juanpablo Heredia, Leighton Estrada-Rayme, Jeremy Matos-Cangalaya, and 1 more author
    In 38th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), 2025
  4. adaptative.png
    Adaptive multimodal emotion detection architecture for social robots
    Juanpablo Heredia, Edmundo Lopes-Silva, Yudith Cardinale, and 4 more authors
    Ieee Access, 2022
  5. multimodal.png
    A multi-modal visual emotion recognition method to instantiate an ontology
    Juanpablo Heredia, Yudith Cardinale, Irvin Dongo, and 1 more author
    In 16th International conference on software technologies, 2021