Skip to main content
ARS Home » Southeast Area » Fayetteville, Arkansas » Poultry Production and Product Safety Research » Research » Publications at this Location » Publication #421638
Research Project: Developing Best Management Practices for Poultry Litter to Improve Agronomic Value and Reduce Air, Soil and Water Pollution

Location: Poultry Production and Product Safety Research

Title: Multi-modal fusion for zero-shot animal detection in UAV-based aerial surveys

Author
item VIRUPAKSHAIAH, ADITI - Mississippi State University
item PILLAI, NISHA - Mississippi State University
item SMITH, HARRISON - University Of Arkansas
item Ashworth, Amanda
item Gowda, Prasanna
item Owens, Phillip
item Rivers, Adam
item NADURI, BINDU - Mississippi State University
item RAMKUMAR, MAHALINGAM - Mississippi State University

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 12/3/2024
Publication Date: 12/3/2024
Citation: Virupakshaiah, A., Pillai, N., Smith, H., Ashworth, A.J., Gowda, P.H., Owens, P.R., Rivers, A.R., Naduri, B., Ramkumar, M. 2024. Multi-modal fusion for zero-shot animal detection in UAV-based aerial surveys. Abstract. Plant and Animal Genome Conference, San Diego, California.

Interpretive Summary:

Technical Abstract: Detecting animals in UAV-based aerial surveys without extensive labeled data is a growing challenge in wildlife monitoring and agricultural management. This study presents a multi-modal fusion framework that combines state-of-the-art models, Grounding DINO and Segment Anything Model (SAM 2), for zero-shot detection and counting of cattle in high-resolution RGB images captured by an Unmanned Aerial Vehicle (UAV). The approach efficiently identifies animals even in complex environments with variations in scale, lighting, and background. By processing high-resolution aerial imagery, the framework achieves accurate results without requiring prior task-specific training. This work highlights the potential of advanced AI techniques in enabling scalable and automated aerial surveys for ecological and agricultural applications.