SOCRATES: Introducing Depth in Visual Wildlife Monitoring Using Stereo Vision

Area of Science:

• Ecology
• Wildlife Biology
• Conservation Technology

Background:

• Efficient species monitoring is crucial for understanding ecosystem changes and population dynamics.
• Camera trapping is valuable for wildlife abundance estimation, but 3D habitat data can enhance accuracy.
• Modern technology integration is key for advancing ecological research and biodiversity assessment.

Purpose of the Study:

• To introduce a novel 3D camera trapping approach, SOCRATES (StereO CameRA Trap for monitoring of biodivErSity).
• To evaluate the effectiveness of SOCRATES in improving wildlife detection and abundance estimation.
• To provide an open-source software and documentation for the SOCRATES system.

Main Methods:

• Development of optimized hardware and software for 3D camera trapping.
• Utilizing stereo vision to capture and infer 3D habitat information.
• Comprehensive evaluation using metrics like bounding box mAP75 and camera trap distance sampling.

Main Results:

• SOCRATES demonstrated a 3.23% improvement in animal detection accuracy (bounding box mAP75).
• The system showed superior applicability for estimating animal abundance via camera trap distance sampling.
• The developed software and documentation are publicly available.

Conclusions:

• SOCRATES offers an advanced 3D camera trapping solution for biodiversity monitoring.
• The integration of 3D data significantly enhances wildlife detection and abundance estimation.
• Open-source availability promotes wider adoption and advancement in ecological research technology.