# SATSN: A Spatial-Adaptive Two-Stream Network for Automatic Detection of Giraffe Daily Behaviors

## TL;DR

A new AI system detects giraffe behaviors like licking and walking to help zoos monitor their health and well-being more effectively.

## Contribution

A novel spatial-adaptive two-stream network (SATSN) is introduced for accurate giraffe behavior detection.

## Key findings

- The SATSN method achieved a mean average precision (mAP) of 93.99% for detecting giraffe behaviors.
- The system uses a video transformer encoder and temporal attention module to improve motion feature representation.
- A dataset of 420 video clips was created and used to validate the method's effectiveness.

## Abstract

We propose a deep learning-based method for detecting daily behaviors in giraffes, aiming to improve detection accuracy and support zoo staff in more effectively monitoring giraffe behavior. This approach promotes a more scientific and proactive strategy for monitoring giraffe health and enhancing their physical and psychological well-being. A spatial-adaptive two-stream network is proposed to reduce false positives and missed detections. Experimental results demonstrate that the method achieves high accuracy and stability in detecting giraffe daily behaviors, offering an effective technological solution for long-term, non-contact, and intelligent behavior monitoring with promising application potential.

The daily behavioral patterns of giraffes reflect their health status and well-being. Behaviors such as licking, walking, standing, and eating are not only essential components of giraffes’ routine activities but also serve as potential indicators of their mental and physiological conditions. This is particularly relevant in captive environments such as zoos, where certain repetitive behaviors may signal underlying well-being concerns. Therefore, developing an efficient and accurate automated behavior detection system is of great importance for scientific management and welfare improvement. This study proposes a multi-behavior automatic detection method for giraffes based on YOLO11-Pose and the spatial-adaptive two-stream network (SATSN). Firstly, YOLO11-Pose is employed to detect giraffes and estimate the keypoints of their mouths. Observation-Centric SORT (OC-SORT) is then used to track individual giraffes across frames, ensuring temporal identity consistency based on the keypoint positions estimated by YOLO11-Pose. In the SATSN, we propose a region-of-interest extraction strategy for licking behavior to extract local motion features and perform daily behavior classification. In this network, the original 3D ResNet backbone in the slow pathway is replaced with a video transformer encoder to enhance global spatiotemporal modeling, while a Temporal Attention (TA) module is embedded in the fast pathway to improve the representation of fast motion features. To validate the effectiveness of the proposed method, a giraffe behavior dataset consisting of 420 video clips (10 s per clip) was constructed, with 336 clips used for training and 84 for validation. Experimental results show that for the detection tasks of licking, walking, standing, and eating behaviors, the proposed method achieves a mean average precision (mAP) of 93.99%. This demonstrates the strong detection performance and generalization capability of the approach, providing robust support for automated multi-behavior detection and well-being assessment of giraffes. It also lays a technical foundation for building intelligent behavioral monitoring systems in zoos.