About FaunaScope

AI-Driven Infrastructure for Wildlife Monitoring

FaunaScope connects computer science and ecology to make wildlife observation more scalable, more intelligent, and more useful for real biological discovery.

Edge + Cloud
Integrated architecture for remote field deployments
Multimodal AI
Automated species, behavior, and interaction annotations
Open Source
Tools designed to support the wider research community

Project Overview

FaunaScope is an interdisciplinary research initiative building next-generation infrastructure for AI-driven wildlife monitoring. The project focuses on both practical challenges at human-wildlife interfaces and a broader biological mission: understanding behavior and interactions in natural environments.

Traditional camera trap workflows often miss nuanced behaviors, inter-species interactions, and long-term patterns. FaunaScope is designed to close that gap and enable richer ecosystem insight at scale.

The Challenge

Field biology now faces a major data and infrastructure bottleneck.
  • Remote deployments with strict power and internet limits
  • High-quality video transfer is often costly or infeasible
  • Manual annotation is too slow for large ecological studies

How FaunaScope Solves It

1. Intelligent Edge Computing

Edge devices (e.g., NVIDIA Jetson Orin) process footage in real time, filtering irrelevant frames and prioritizing biologically meaningful events.

2. Advanced Compression & Transmission

Learned neural video codecs and aggressive compression make low-bandwidth transmission practical, with cloud-side reconstruction for analysis.

3. Automated Analysis with Multimodal AI

Cloud multimodal models annotate species, behaviors, interactions, and environmental context, enabling semantic search across large video archives.

4. Autonomous Field Infrastructure

Weather-resistant hardware with LiFePO4 batteries, solar generation, and smart power management supports 24/7, unattended operation.

Open Source Commitment

FaunaScope software is released as open source to support the broader scientific and conservation community.

Edge Repository: github.com/imics-lab/faunascope-edge

Funding & Lab

This work is supported by the Texas State University Research Enhancement Program (REP).

Research is conducted in the Intelligent Multimodal Computing and Sensing (IMICS) Lab at Texas State University.

Research Team

Principal Investigators
  • Dr. Vangelis Metsis (PI): Associate Professor, Computer Science
  • Dr. Ivan Castro-Arellano (Co-PI): Associate Professor, Biology
  • Dr. Joseph Veech (Co-PI): Professor, Biology
Student Researchers
  • Mykhailo Sakevych — Ph.D. Student, Computer Science
  • Felix Mathew — M.S. Student, Computer Science
  • Himson Chapagain — Undergraduate Student, Computer Science
  • Brianna Gutierrez — M.S. Student, Wildlife Ecology