Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.10923 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913864827797504 |
|---|---|
| author | Adebola, Simeon Xie, Shuangyu Kim, Chung Min Kerr, Justin van Marrewijk, Bart M. van Vlaardingen, Mieke van Daalen, Tim van Loo, E. N. Rincon, Jose Luis Susa Solowjow, Eugen van de Zedde, Rick Goldberg, Ken |
| author_facet | Adebola, Simeon Xie, Shuangyu Kim, Chung Min Kerr, Justin van Marrewijk, Bart M. van Vlaardingen, Mieke van Daalen, Tim van Loo, E. N. Rincon, Jose Luis Susa Solowjow, Eugen van de Zedde, Rick Goldberg, Ken |
| contents | Accurate temporal reconstructions of plant growth are essential for plant phenotyping and breeding, yet remain challenging due to complex geometries, occlusions, and non-rigid deformations of plants. We present a novel framework for building temporal digital twins of plants by combining 3D Gaussian Splatting with a robust sample alignment pipeline. Our method begins by reconstructing Gaussian Splats from multi-view camera data, then leverages a two-stage registration approach: coarse alignment through feature-based matching and Fast Global Registration, followed by fine alignment with Iterative Closest Point. This pipeline yields a consistent 4D model of plant development in discrete time steps. We evaluate the approach on data from the Netherlands Plant Eco-phenotyping Center, demonstrating detailed temporal reconstructions of Sequoia and Quinoa species. Videos and Images can be seen at https://berkeleyautomation.github.io/GrowSplat/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_10923 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GrowSplat: Constructing Temporal Digital Twins of Plants with Gaussian Splats Adebola, Simeon Xie, Shuangyu Kim, Chung Min Kerr, Justin van Marrewijk, Bart M. van Vlaardingen, Mieke van Daalen, Tim van Loo, E. N. Rincon, Jose Luis Susa Solowjow, Eugen van de Zedde, Rick Goldberg, Ken Robotics Computer Vision and Pattern Recognition Accurate temporal reconstructions of plant growth are essential for plant phenotyping and breeding, yet remain challenging due to complex geometries, occlusions, and non-rigid deformations of plants. We present a novel framework for building temporal digital twins of plants by combining 3D Gaussian Splatting with a robust sample alignment pipeline. Our method begins by reconstructing Gaussian Splats from multi-view camera data, then leverages a two-stage registration approach: coarse alignment through feature-based matching and Fast Global Registration, followed by fine alignment with Iterative Closest Point. This pipeline yields a consistent 4D model of plant development in discrete time steps. We evaluate the approach on data from the Netherlands Plant Eco-phenotyping Center, demonstrating detailed temporal reconstructions of Sequoia and Quinoa species. Videos and Images can be seen at https://berkeleyautomation.github.io/GrowSplat/ |
| title | GrowSplat: Constructing Temporal Digital Twins of Plants with Gaussian Splats |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.10923 |