A5022120434- Human Pose and Action Recognition
- Remote Sensing in Agriculture
- Animal Vocal Communication and Behavior
- Wildlife-Road Interactions and Conservation
- Anomaly Detection Techniques and Applications
- Water Quality Monitoring Technologies
- Wildlife Ecology and Conservation
- Computational Geometry and Mesh Generation
- AI in cancer detection
- Image Processing and 3D Reconstruction
- Video Analysis and Summarization
- Data Visualization and Analytics
- Computer Graphics and Visualization Techniques
- Machine Learning in Healthcare
- Artificial Intelligence in Healthcare
- Advanced Image and Video Retrieval Techniques
- Time Series Analysis and Forecasting
- Animal Behavior and Welfare Studies
- Remote Sensing and LiDAR Applications
- Digital Media Forensic Detection
- Sports Analytics and Performance
- Underwater Acoustics Research
- Bayesian Modeling and Causal Inference
University of Auckland
2023-2024
Universidad Nacional Autónoma de México
2020
The application of Artificial Intelligence (AI) and Computer Vision (CV) in sports has generated significant interest enhancing viewer experience through graphical overlays predictive analytics, as well providing valuable insights to coaches. However, more efficient methods are needed that can be applied across different without incurring high data annotation or model training costs. A major limitation deep learning models on large datasets is the resource requirement for reproducing...
Deep learning approaches for animal re-identification have had a major impact on conservation, significantly reducing the time required many downstream tasks, such as well-being monitoring. We propose method called Recurrence over Video Frames (RoVF), which uses recurrent head based Perceiver architecture to iteratively construct an embedding from video clip. RoVF is trained using triplet loss co-occurrence of individuals in frames, where individual IDs are unavailable. tested this and...
Recording animal behaviour is an important step in evaluating the well-being of animals and further understanding natural world. Current methods for documenting within a zoo setting, such as scan sampling, require excessive human effort, are unfit around-the-clock monitoring, may produce human-biased results. Several datasets already exist that focus predominantly on wildlife interactions, with some extending to action or recognition. However, there limited data setting focusing group...
Better understanding the natural world is a crucial task with wide range of applications. In environments close proximity between humans and animals, such as zoos, it essential to better understand causes behind animal behaviour what interventions are responsible for changes in their behaviours. This can help predict unusual behaviours, mitigate detrimental effects increase well-being animals. There has been work on modelling dynamics swarms birds insects but complex social behaviours...