Realistic Procedural Plant Modeling Guided by 3D Point Cloud


In Proceedings of SIGGRAPH ’17 Posters


Jianwei Guo      Zhanglin Cheng2       Shibiao Xu3       Xiaopeng Zhang4

    

1NLPR, Institute of Automation, CAS       2Shenzhen VisuCA Key Lab, SIAT, CAS      3NLPR, Institute of Automation, CAS      4NLPR, Institute of Automation, CAS    



Figure 1: Modeling of a small scene from street-level scanned data. The images show a photo of the scene (top left), point cloud (top right) and generated plant models with textured leaves (bottom).


Abstract


Plants are ubiquitous in the nature, and realistic plant modeling plays an important role in a variety of applications. Over the last decades, an immense amount of eorts have been dedicated to plant modeling. These approaches can be classied into two major categories: procedural modeling and data-driven reconstruction approaches. In this paper, we present a novel modeling framework for generating realistic plants by integrating point cloud analysis with rule-based growth for procedural plant modeling. Our method makes several important contributions to the research on plant modeling:

    • enriching the tree generation literature by building connections between virtual tree modeling and data-driven tree reconstruction;

    • generating ground-covering non-tree plants, e.g., bushes and shrubs, which have gained little attention and cannot be reconstructed by previous approaches.



OverView





Figure 2:Simulating the growing process of a tree.


Figure 3:We show the color-coded weight image (warmer color indicates higher weight), as well as modeling results without/with the weight.


Results

We present a rule-based framework for generating naturally-looking plant models from real point cloud. In the future, we would like to address automatic classification and modeling of larger scale scenes, such as a real ecosystem.


Acknowledgements

This work is supported in part by the National Natural Science Foundation of China (Nos. 61331018, 61379091, 61372168 and 61571439).

Copyright © 1998-2018 Visual Computing Research Center