Special Session: Data in Agriculture
Foundation Models for Wheat: A Data-Centric Odyssey
- Venue: Bali (in-person)
- Bali Time: Fri, 5 Dec 2025 15:00 - 16:00 WITA (UTC+8)
Vision-driven field monitoring is central to digital agriculture, yet models built on general-domain pretrained backbones often fail to generalize across tasks, owing to the interaction of fine, variable canopy structures with fluctuating field conditions. We present FoMo4Wheat, one of the first crop-domain vision foundation model pretrained with self-supervision on ImAg4Wheat, the largest and most diverse wheat image dataset to date (2.5 million high-resolution images collected over a decade at 30 global sites, spanning >2,000 genotypes and >500 environmental conditions). This wheat-specific pretraining yields representations that are robust for wheat and transferable to other crops and weeds. Across ten in-field vision tasks at canopy and organ levels, FoMo4Wheat models consistently outperform state-of-the-art models pretrained on general-domain dataset. These results demonstrate the value of crop-specific foundation models for reliable in-field perception and chart a path toward a universal crop foundation model with cross-species and cross-task capabilities.
Speaker
Hao Lu
Huazhong University of Science and Technology
Hao Lu is an Associate Professor with Huazhong University of Science and Technology, Wuhan, China. He has over ten years research experience in the intersected field of computer vision and plant phenotyping. He is Associate Editor of Plant Phenomics, a Senior Program Committee Member of AAAI 2026, (Executive) Area Chair of VALSE and PRCV 2023. He received the EurAgEng Outstanding Paper Award (2018), the Top Performance Award at MLCAS/EPPS/ICCV 2025 Global Wheat Full Semantic Segmentation Challenge, Runner-Up Award at ICCV 2021 Global Wheat Head Detection Challenge, and the Bronze Medal at the International Exhibition of Inventions Geneva 2018. He is a Senior Member of IEEE.
Industry Session
Engineering PhoebeDB for a Cost-Efficient Future
- Venue: Sydney (in-person)
- Sydney Time: Sat, 6 Dec 2025 10:00 - 11:00 AEDT (UTC+11)
Speaker
Yi Jin
Enmotech Data AU
Yi Jin is the co-founder and Chief Architect of the PhoebeDB project, where he leads the engineering of next-generation high-performance commercial database systems. He also serves as an Adjunct Lecturer at UNSW. Dr. Jin received his Ph.D. in Computer Software and Theory in 2009 and has since accumulated more than 15 years of in-depth experience in the database industry. His career spans enterprise-grade product research and development, with hands-on contributions across the full DBMS stack—from storage engines and query processing to distributed systems and performance optimization. He brings a unique blend of academic rigor and real-world engineering insight to modern database system design.