WildAI is currently in the final stages of development.
Use our AI model to automatically detect large mammals in your aerial photographs and get accurate counts and species identification.
Supported By


Uliege.jpg)
An initiative of ULiège Gembloux Agro-Bio Tech
Uliege.jpg)
WildAI is a project developed by researchers from the University of Liège, Gembloux Agro-Bio Tech (Belgium). It is the result of years of research on new methods for wildlife population surveys and benefits from close collaborations with various field actors across Africa.
Researchers and conservation enthusiasts
We are a team of researchers and conservation enthusiasts working to improve wildlife population survey methods for large African mammals. By combining aerial imagery and artificial intelligence, we are developing an innovative solution to automate and optimize the analysis of wildlife counting data.

Dr. Alexandre
Delplanque
Co-Founder
Project Manager
AI Developer

Prof. Philippe
Lejeune
Co-Founder
Project Manager

Nicolas
Delplanque
UI/UX Designer
Full-Stack Developer

Ir. Hugues
Dethier
Data Management
Application Reviewer
Alexandre_Delplanque.jpg)
Enhance aerial survey methods with remote sensing and AI
Alexandre_Delplanque.jpg)
Traditional wildlife surveys conducted from light aircraft are expensive, risky, and sometimes inaccurate. Using aerial photography combined with AI models allows:
- More accuratepopulation estimates
- Reduced costs and human risks
- Significanttime savings
Provide powerful tools for research and conservation
Alexandre_Delplanque.jpg)
We believe in a modern, technology-driven approach to better monitor and protect biodiversity. Our goal is to provide researchers, protected area managers, and conservationists with a powerful tool to easily process images from their aerial surveys.
Alexandre_Delplanque.jpg)
Alexandre_Delplanque.jpg)
A free and collaborative platform
Alexandre_Delplanque.jpg)
WildAI is completely free* to use. In exchange for this service, users agree that their uploaded images and verified detections are securely stored and used to enhance the AI model.
* Up to a maximum of 500 processed images.