Grow with the Flow: 4D Reconstruction of Growing Plants with Gaussian Flow Fields

2026-02-09

AI Summary

This work addresses the challenge of modeling dynamic plant growth, where existing methods struggle to effectively capture the continuous emergence of new geometric structures. We propose a 4D reconstruction approach based on a 3D Gaussian flow field, representing plant growth as the continuous nonlinear evolution of Gaussian parameters—including position, scale, orientation, color, and opacity—over time. To accurately model morphological changes across the entire life cycle, we introduce an inverse growth strategy for initializing Gaussian primitives. This is the first application of Gaussian flow fields to plant growth modeling, overcoming key limitations of conventional deformation fields and 4D Gaussian splatting in generating novel geometry and capturing temporal trajectories. Experiments on multi-view time-lapse plant growth datasets demonstrate that our method significantly outperforms state-of-the-art techniques in both image fidelity and geometric accuracy.

Technology Category

• Computer Vision: 3D Computer Vision
• Search and Optimization: Mixed Discrete/Continuous Search
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Modeling the time-varying 3D appearance of plants during their growth poses unique challenges: unlike many dynamic scenes, plants generate new geometry over time as they expand, branch, and differentiate. Recent motion modeling techniques are ill-suited to this problem setting. For example, deformation fields cannot introduce new geometry, and 4D Gaussian splatting constrains motion to a linear trajectory in space and time and cannot track the same set of Gaussians over time. Here, we introduce a 3D Gaussian flow field representation that models plant growth as a time-varying derivative over Gaussian parameters -- position, scale, orientation, color, and opacity -- enabling nonlinear and continuous-time growth dynamics. To initialize a sufficient set of Gaussian primitives, we reconstruct the mature plant and learn a process of reverse growth, effectively simulating the plant's developmental history in reverse. Our approach achieves superior image quality and geometric accuracy compared to prior methods on multi-view timelapse datasets of plant growth, providing a new approach for appearance modeling of growing 3D structures.

Problem

• 4D reconstruction
• plant growth
• time-varying 3D appearance
• new geometry generation
• dynamic scenes

Innovation

• Gaussian Flow Fields
• 4D Reconstruction
• Plant Growth Modeling
• Nonlinear Temporal Dynamics
• Reverse Growth Simulation