International interest in Purdue AI research takes root at 2025 Venice Biennale

07-11-2025

Tree-D Fusion model displayed at Venice Biennale

An artificial intelligence system developed by Jae Joong Lee, a PhD student in Purdue University’s Department of Computer Science, has been selected for exhibition at the 2025 Venice Biennale of Architecture, one of the world’s most prestigious showcases for design, sustainability and innovation.

The project, titled Tree-D Fusion, creates three-dimensional models of real trees using just a single photograph—transforming flat images into digital twins that reveal a tree’s past growth and predict its future development. The AI-driven model offers real-time insights into how a tree’s canopy will change over time, how much shade it will cast throughout the year, and how it interacts with the environment.

Lee led the development of Tree-D Fusion. His research combines deep learning, computer vision and graphics to build digital replicas of real-world objects and ecosystems. The opportunity to feature the work in the Biennale came through a collaboration with MIT’s Senseable City Lab. Researchers at the lab, including MIT assistant professor Sara Beery, contacted Lee after seeing his previous work on digital tree models. Together, the team submitted a proposal that was accepted by the Biennale’s curators.

“This is more than a visualization tool—it’s a way of understanding and interacting with nature,” Lee said. “Tree-D Fusion takes one image and reconstructs a tree’s life story. It’s a scientific tool with real-world uses, but also something that speaks to creativity and design.”

Tree-D Fusion generates 3D tree models that allow researchers, designers and planners to simulate how a tree will grow and interact with its surroundings. The system can estimate the amount of shade a tree will cast in the future, making it a useful tool for urban planning, landscape architecture, forestry and even solar energy placement. It has the potential to eliminate costly LiDAR surveys and enable fast, accessible tree modeling using only a smartphone photo.

“Think of Tree-D Fusion like a single frame in a time-lapse video,” Lee said. “Even with one photo, the AI can recreate what the tree looked like in the past and project how it will grow, filling in the details that are hidden or missing.”

Lee’s work has drawn attention within the academic community for its technical strength. Tree-D Fusion appeared at European Conference on Computer Vision (ECCV) 2024, a leading computer vision conference. Related work, including Latent L-systems, a deep-learning model for generating tree structures, was accepted at SIGGRAPH 2024. His most recent work, RGB2Point, which converts a single image into a 3D point cloud, will be presented at the Winter Conference on Applications of Computer Vision (WACV) 2025.

The inclusion of Tree-D Fusion in the Venice Biennale highlights the significance of Lee’s work at the intersection of science, technology and art.

“Science, technology, and art are deeply interconnected and each drives progress in the others. Artistic creativity can spark scientific inquiry, and scientific discoveries often inspire new forms of artistic expression,” says Lee.

“Technological innovations provide tools that enhance both research and artistic practice. This reciprocal relationship fuels advancement across all three domains, and precisely this synergy makes me passionate about each of them.

Looking ahead, Lee aims to expand his research beyond trees to create digital twins of crops and plants. His long-term goal is to use AI to forecast growth, improve agricultural practices and help build sustainable systems that respond to climate change.

The project was developed through a collaboration between several researchers at Purdue. Lee led the design and implementation of the system. Bosheng Li, who completed his Ph.D. at Purdue, contributed to the tree growth simulation. Bedrich Benes, professor and associate head of computer science, provided overall project leadership and expertise in computer graphics. Raymond Yeh, assistant professor of computer science, led the AI and computer vision research. Songlin Fei, professor and chair of remote sensing and director of Purdue’s Institute for Digital Forestry, supported the project through his expertise and institutional leadership.

Outside of Purdue, collaborators included Beery, who contributed insight on tree datasets, and Jonathan Huang, formerly of Google and now head of AI at General Robotics. Huang provided technical expertise and access to key datasets.

The project was supported by Purdue’s high-performance computing infrastructure through the Rosen Center for Advanced Computing and by a collaborative research culture across departments. Lee conducts his work through the Institute for Digital Forestry and the Computer Graphics and Visualization Laboratory.

Funding for Tree-D Fusion came from multiple sources, including the USDA Natural Resources Conservation Service through the EFFICACI grant, the USDA National Institute of Food and Agriculture grant titled “Promoting Economic Resilience and Sustainability of the Eastern U.S. Forests,” and a NIFA-AFRI award aimed at improving forest management through automated measurement of tree geometry.

To view images of the display at the 2025 Venice Biennale, click here.

To see the interactive ReLeaf website, click here.

 

About the Department of Computer Science at Purdue University

Founded in 1962, the Department of Computer Science was created to be an innovative base of knowledge in the emerging field of computing as the first degree-awarding program in the United States. The department continues to advance the computer science industry through research. U.S. News & World Report ranks the department No. 8 in computer engineering and No. 16 overall in undergraduate and graduate computer science. Additionally, the program is ranked No. 6 in cybersecurity, No. 8 in software engineering, No. 13 in systems, No. 15 in programming languages and data analytics, and No. 18 in theory. Graduates of the program are able to solve complex and challenging problems in many fields. Our consistent success in an ever-changing landscape is reflected in the record undergraduate enrollment, increased faculty hiring, innovative research projects, and the creation of new academic programs. The increasing centrality of computer science in society, academic disciplines and new research activities — centered around foundations and applications of artificial intelligence and machine learning, such as natural language processing, human computer interaction, vision, and robotics, as well as systems and security — are the future focus of the department. Learn more at cs.purdue.edu.