Abstract

This project visualizes the collaborative relationships between institutions engaged in neuromorphic vision research. By analyzing co-authorship patterns from academic papers, we create an interactive world map that shows the global distribution of neuromorphic vision research institutions and their collaboration networks. The visualization reveals the interconnected nature of this emerging field and highlights major research hubs worldwide.

🗺️ View Interactive Map


Project Overview

The neuromorphic vision field has seen tremendous growth in recent years, with researchers worldwide contributing to advances in event-based cameras, spiking neural networks, and bio-inspired vision algorithms. This visualization project aims to map the global landscape of neuromorphic vision research by analyzing institutional collaborations through academic publications.

The interactive map displays:

  • Geographic distribution of research institutions
  • Collaboration networks based on co-authorship patterns
  • Institution details and research focus areas
  • Publication statistics for each institution


Data Sources and Methodology

Data Collection

The paper list used in this project is sourced from a comprehensive Google Sheets database, with data collected up to August 19, 2025. Based on the paper titles and conference/journal information in the list, I used AI agents to query author lists through arXiv and Semantic Scholar APIs.

Institution Extraction

For papers available on arXiv, I employed AI agents to extract and clean institutional information from their front pages, followed by manual verification and additional cleaning work. The geographic coordinates used to visualize institutional locations are provided by large language models.

Data Limitations

Due to technical limitations, information for many papers was not successfully extracted, and many authors’ institutions remain unannotated (categorized as “unknown”). The positions on the map are not precise and are adjusted using a simple algorithm to spread out overly dense points for better visualization.

Technical Implementation

The visualization webpage code was primarily written by Claude 4-based Copilot under my guidance through several iterations, combining modern web technologies for interactive data visualization.


Acknowledgments

Thank you to all neuromorphic vision practitioners for their contributions to this rapidly evolving field. I hope we can work together to make the industry flourish and make the world better!

Special thanks to the research community for maintaining open access to publication data and enabling collaborative research visualization projects like this one.

Author: HYLZ (GitHub)
Last Updated: August 25, 2025