RGB-D Perception System for Edge AI
Built and deployed a real-time RGB-D perception pipeline integrating camera sensing, object detection, identity-linked data capture, automated preprocessing, and remote monitoring for large-scale field operation.
Transfer Learning & Depth Image vs. Point Cloud
The objectives of this study were to evaluate whether transfer learning from a large farm enhances body weight prediction on a small farm with limited data, and compare the predictive performance of depth-image- and point-cloud-based approaches.
Wearable sensor-based cattle behavior prediction using ML and DL
This study is aimed to assess the impact of CV designs on the prediction of multi-class imbalanced cattle behavior and compare the prediction performance of ML and DL models using both random and block CV.
An open-source, cloud-based computer vision web application for object detection and segmentation, three-dimensional visualization, as well as 2D and 3D morphological feature extraction.