Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Whole-Herd Elephant Pose Estimation from Drone Data for Collective Behavior Analysis

Brody McNutt, Libby Zhang, Angus Carey-Douglas, Fritz Vollrath, Frank Pope, Leandra Brickson

Abstract

This research represents a pioneering application of automated pose estimation from drone data to study elephant behavior in the wild, utilizing video footage captured from Samburu National Reserve, Kenya. The study evaluates two pose estimation workflows: DeepLabCut, known for its application in laboratory settings and emerging wildlife fieldwork, and YOLO-NAS-Pose, a newly released pose estimation model not previously applied to wildlife behavioral studies. These models are trained to analyze elephant herd behavior, focusing on low-resolution ($\sim$50 pixels) subjects to detect key points such as the head, spine, and ears of multiple elephants within a frame. Both workflows demonstrated acceptable quality of pose estimation on the test set, facilitating the automated detection of basic behaviors crucial for studying elephant herd dynamics. For the metrics selected for pose estimation evaluation on the test set -- root mean square error (RMSE), percentage of correct keypoints (PCK), and object keypoint similarity (OKS) -- the YOLO-NAS-Pose workflow outperformed DeepLabCut. Additionally, YOLO-NAS-Pose exceeded DeepLabCut in object detection evaluation. This approach introduces a novel method for wildlife behavioral research, including the burgeoning field of wildlife drone monitoring, with significant implications for wildlife conservation.

Whole-Herd Elephant Pose Estimation from Drone Data for Collective Behavior Analysis

Abstract

This research represents a pioneering application of automated pose estimation from drone data to study elephant behavior in the wild, utilizing video footage captured from Samburu National Reserve, Kenya. The study evaluates two pose estimation workflows: DeepLabCut, known for its application in laboratory settings and emerging wildlife fieldwork, and YOLO-NAS-Pose, a newly released pose estimation model not previously applied to wildlife behavioral studies. These models are trained to analyze elephant herd behavior, focusing on low-resolution (50 pixels) subjects to detect key points such as the head, spine, and ears of multiple elephants within a frame. Both workflows demonstrated acceptable quality of pose estimation on the test set, facilitating the automated detection of basic behaviors crucial for studying elephant herd dynamics. For the metrics selected for pose estimation evaluation on the test set -- root mean square error (RMSE), percentage of correct keypoints (PCK), and object keypoint similarity (OKS) -- the YOLO-NAS-Pose workflow outperformed DeepLabCut. Additionally, YOLO-NAS-Pose exceeded DeepLabCut in object detection evaluation. This approach introduces a novel method for wildlife behavioral research, including the burgeoning field of wildlife drone monitoring, with significant implications for wildlife conservation.

Paper Structure

This paper contains 13 sections, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Dataset
  4. Preprocessing
  5. DeepLabCut Workflow
  6. Elephant Detector
  7. DeepLabCut
  8. YOLO-NAS-Pose workflow
  9. Evaluation
  10. Video Tracker for Visualization
  11. Results
  12. Discussion
  13. Conclusion

Figures (5)

  • Figure 1: Example frame from captured drone footage from Save the Elephants in Samburu National Reserve, Kenya. The resolution has been greatly reduced for this manuscript.
  • Figure 2: Keypoints of desired pose estimation in order to achieve automated detection for ear flapping, head orientation.
  • Figure 3: Workflow details for both methods investigated
  • Figure 4: Example of pose estimations generated by DeepLabCut on patches extracted from the YOLOv5 detector.
  • Figure 5: Example of a test set image with pose estimations YOLO-NAS-Pose overlaid. Though there is decent performance, with only one false-positive calf, the “forehead” keypoint is consistently off for all detected elephants.