University of Tokyo
Today, efficient, cost-effective sensors and high performance computing technologies are looking to transform traditional plant-based agriculture into an efficient cyber-physical system. The easy availability of cheap, deployable, connected sensor technology has created an enormous opportunity to collect vast amounts of data at varying spatial and temporal scales at both experimental and production agriculture levels. Therefore, both offline and real-time agricultural analytics that assimilate such heterogeneous data and provide automated, actionable information are critical for sustainable and profitable agriculture.
Data analytics and decision-making for Agriculture has been a long-standing application area. The application of advanced Artificial Intelligence (AI) and Machine Learning (ML) methods to this critical societal need can be viewed as a transformative extension for the agriculture community. In this workshop, we intend to bring together academic and industrial researchers and practitioners in the fields of machine learning, data science and engineering, plant sciences and agriculture, in the collaborative effort of identifying and discussing major technical challenges and recent results related to machine learning-based approaches. It will feature invited talks, oral/poster presentations of accepted papers, and an Ag-ML competition.
Category | Before August 1 | After August 1 (onside payment only) |
---|---|---|
Industry Professional | ¥15,000 | ¥20,000 |
Academia/Non-Profit/Start-Up | ¥10,000 | ¥15,000 |
Students | ¥3,000 | ¥5,000 |
The Seventh International Workshop on Machine Learning for Cyber-Agricultural Systems (MLCAS) 2025 will focus on the theme “Convergence of Multi-Modal Sensing, Multi-Omics Integration, and Multi-Platform Analytics for Next-Generation Cyber-Agricultural Systems.” This year’s workshop aims to explore how synergizing heterogeneous data streams—from spectral imaging and IoT sensor networks to genomic and environmental datasets—can revolutionize precision agriculture. By bridging advances in machine learning with multi-scale biological, environmental, and operational data, MLCAS2025 will address critical challenges in crop resilience, resource optimization, and climate-smart farming. The workshop will emphasize computational frameworks capable of harmonizing data across spatial, temporal, and biological scales, enabling predictive digital twins that integrate plant physiology, field conditions, and management practices.
The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support..
Click to swtich days
Time(JST) |
Chair |
Title |
Presenter |
---|---|---|---|
09:00~09:20 | Dr. Wei GUO | Opening remarks and Introduction | Dr. Masayuki Hirafuji |
09:20~09:30 | Welcome message | Dr. Soumik Sarkar | |
09:30~10:00 | Invited Talk 1 Root Phenomics - Unraveling unknown functional phenotypes from cells to populations |
Dr. Alexander Bucksch | |
10:00~10:30 | Invited Talk 2 Data-Centric smart AI for high-throughput plant phenotyping |
Dr. Haiyan Cen | |
10:30~11:00 | Coffee break | ||
11:00~11:15 | Dr. Ian Stavness | Digital Transformation of the Forestry Industry in Japan | James Burridge |
11:15~11:30 | Artificial Intelligence Organ-Level Identification of Greenhouse-Growth Cucumis melo using Visible images and Near-infrared Hyperspectral Images | Yuan-Kai Tu | |
11:30~11:45 | An unsupervised 3D point cloud segmentation method and tool for artificial objects and trees in dense orchards | Tianhai Wang | |
11:45~12:00 | Soybean Yield Prediction with Bidirectional Linear Unit for Recurrence | Qisai Liu | |
12:00~13:00 | Lunch break | ||
13:00~13:30 | Dr. Soumik Sarkar | Invited Talk 3 Harnessing Data-Driven Breeding for Climate-Resilient Agriculture |
Dr. Hiroyoshi Iwata |
13:30~14:00 | Invited Talk 4 AI in agriculture — from disease identification to question answering — using tomato as an example |
Dr. Yan-Fu Kou | |
14:00~14:15 | WeedNet: A Foundation Model-Based Global-to-Local AI Approach for Real-Time Weed Species Identification and Classification | Ashlyn Rairdin | |
14:15~14:30 | Camera Align and Folder Drag are All You Need: Rapid Crop Lodging Aerial Assessment without Segmentation | Haozhou Wang | |
14:30~14:50 | Coffee break | ||
14:50~15:05 | Dr. Masayuki Hirafuji | Physically Consistent Hyperspectral NeRF for 3D Plant Phenotyping using a Hyperspectral Camera | Kibon Ku |
15:20~15:35 | Towards Large Reasoning Models for Agriculture | Elizabeth Tranel & Ashlyn Rairdin | |
15:35~15:50 | AgroBench: Vision-Language Model Benchmark in Agriculture | Risa Shinoda | |
15:50~16:05 | AgAdvisor: Agentic AI for agricultural query resolution | Joshua R. Waite | |
16:05~16:20 | Phylogenetic and Structural Modeling for Optimization of Agricultural Production | José L. J. Ruelas | |
16:20~16:35 | Fine-Tuning and Deploying Vision-Language-Action Models for Visual Servoing with Soft Continuum Arms | Hsin-Jung Yang | |
16:50~17:05 | Mechanics-Informed Phenotyping of Maize Stalks Using Octree-Based Elastic Simulations | Samundra Karki | |
17:30~19:30 | Welcome reception at UTokyo Yayoi Campus |
Time(JST) |
Chair |
Title |
Presenter |
---|---|---|---|
09:00~09:30 | Dr. Baskar Ganapathysubramanian | Invited Talk 5 FOMO4Wheat: Foundation Model for Multi-Task wheat phenotyping |
Dr. Shouyang Liu |
09:30~10:00 | Invited Talk 6 Plant Phenomics to Plant Image Science |
Dr. Do-Soon Kim | |
10:00~10:30 | Coffee break | ||
10:30~12:00 | Dr. Sachiko Isobe & Dr. Alexander Bucksch |
Panel Discussion The Science of Phenomics |
Dr. Yanfeng Ding, Dr. YuanKai Tu, Dr. Hyoung Seok Kim, Dr. Fumio Okura |
12:00~13:00 | Lunch break | ||
13:00~14:00 | Dr. Ian Stavness | Competition Presentations and Award Ceremony | Dr. Zijian Wang |
14:00~14:15 | Find the Fruit: Designing a Zero-Shot Sim2Real Deep RL Planner for Occlusion Aware Plant Manipulation | Nitesh Subedi | |
14:15~14:35 | Coffee break | ||
14:35~14:50 | Dr. Seishi Ninomiya | GaussianPlant: Structure-aligned Gaussian Splatting for 3D Reconstruction of Plants | Fumio Okura & Yang Yang |
14:50~15:05 | Near-field abiotic stress tolerance system for crop stress induction, phenotyping, and breeding applications | Heidi M. Dornath | |
15:05~15:20 | Plant stem skeletonization with Gaussian splat guided contraction | Ian Stavness | |
15:20~15:35 | PlantPose: Plant Skeleton Estimation in 2D and 3D | Xinpeng Liu | |
15:35~15:50 | Sewing the future of cotton: a multi-omics study combining nanomechanics, transcriptomics, and phenotypic traits | Anwesha Sarkar | |
15:50~16:05 | Phenotyping pipeline utilizing DeeplabV3+ for cyber physical systems: A case study of potato crops | Stephen Njehia Njane | |
16:05~16:20 | Dr. Wei Guo | Imputing Spatio-Temporal Gaps in Sentinel-1/2 Imagery with MODIS-Conditioned Deep Models for Downstream Ag Management and Yield Prediction | Bitgoeul Kim |
16:20~16:35 | NeuraLeaf: Neural Parametric Leaf Models with Shape and Deformation Disentanglement | Yang Yang & Fumio Okura | |
16:35~16:50 | Immersive Virtual Reality (VR) Visualization of 3D Models of Millet Plants Reconstructed using Neural Radiance Fields (NeRFs) | Shambhavi Joshi | |
17:05~17:15 | Dr. Wei Guo | ENDING | Dr. Soumik Sarkar |
Global Wheat Full Semantic Segmentation
The details for participation can be found here.
To be eligible for the prizes, participants will have to release the code to their solutions under an open source license of their choice and agree for a post-competition presentation and interview. The submitted code must be reproducible and produce the same score as on the leaderboard. Winners shall, as a condition for receiving their prize, grant to the Organizer a perpetual, worldwide, non-exclusive, royalty-free, transferable, irrevocable license to use their Submission Materials (and any intellectual property relating thereto) for any purpose, including the right to reproduce, modify, prepare derivative works, publicly display, sublicense, and distribute them. The winning participants will have to provide a valid and unexpired ID card or passport which the organizer will use only for the purpose of verifying the individual's identity in case of travel reimbursement and for internal record keeping. Any and all prize(s) is(are) non transferable. All taxes, fees, and expenses associated with participation in the Challenge or receipt and use of a prize are the sole responsibility of the Prize Winner(s). No substitution of prize or transfer/assignment of prize to others or request for the cash equivalent by winners is permitted. Acceptance of prize constitutes permission for the Organizers to use the winner’s name and entry for purposes of advertising and trade without further compensation unless prohibited by law.
For details regarding the competition, please contact us:
Webpage managed by Haozhou Wang, University of Tokyo. For any concerns please contact haozhou-wang[at]g.ecc.u-tokyo.ac.jp
Powered by Github-Pages and Bootstrap4