Branching Gaussian Processes with Applications to Spatiotemporal Reconstruction of 3D Trees

  title={Branching Gaussian Processes with Applications to Spatiotemporal Reconstruction of 3D Trees},
  author={Kyle Simek and Ravishankar Palanivelu and Kobus Barnard},
We propose a robust method for estimating dynamic 3D curvilinear branching structure from monocular images. While 3D reconstruction from images has been widely studied, estimating thin structure has received less attention. This problem becomes more challenging in the presence of camera error, scene motion, and a constraint that curves are attached in a branching structure. We propose a new general-purpose prior, a branching Gaussian processes (BGP), that models spatial smoothness and temporal… 
3D Phenotyping of Plants
This chapter presents a broad overview of computer vision based 3D plant Phenotyping techniques, and some open challenges of vision-based plant phenotyping are discussed, followed by conclusion and some hands on exercises.
BGP: Branched Gaussian processes for identifying gene-specific branching dynamics in single cell data
The branching Gaussian process (BGP) is developed, a non-parametric model that is able to identify branching dynamics for individual genes and provides an estimate of branching times for each gene with an associated credible region.
BGP: identifying gene-specific branching dynamics from single-cell data with a branching Gaussian process
The branching Gaussian process (BGP) is developed, a non-parametric model that is able to identify branching dynamics for individual genes and provide an estimate of branching times for each gene with an associated credible region.
Computer Vision Problems in 3D Plant Phenotyping
A novel system for automated and non-invasive/non-contact plant growth measurement, exploit the recent advancements of sophisticated robotic technologies and near infrared laser scanners to build a 3D imaging system and use state-of-the-art Computer Vision algorithms to fully automate growth measurement.


Automated reconstruction of tree structures using path classifiers and Mixed Integer Programming
This paper formulate the delineation problem as one of solving a Quadratic Mixed Integer Program (Q-MIP) in a graph of potential paths, which can be done optimally up to a very small tolerance, and proposes a novel approach to weighting these paths which results in a Q-Mip solution that accurately matches the ground truth.
Reconstructing Evolving Tree Structures in Time Lapse Sequences
This work proposes an approach to reconstructing tree structures that evolve over time in 2D images and 3D image stacks such as neuronal axons or plant branches and shows that this problem can be formulated as a Quadratic Mixed Integer Program and solved efficiently.
Automated Reconstruction of Dendritic and Axonal Trees by Global Optimization with Geometric Priors
A novel probabilistic approach to fully automated delineation of tree structures in noisy 2D images and 3D image stacks that uses the DIADEM metric to quantitatively evaluate the topological accuracy of the reconstructions and showed that the use of the geometric regularization yields a substantial improvement.
Robust non-rigid registration of 2D and 3D graphs
In the absence of appearance information, this approach iteratively establish correspondences between graph nodes, update the structure accordingly, and use the current mapping estimate to find the most likely correspondences that will be used in the next iteration, which makes the computation tractable.
Modeling and generating moving trees from video
This work presents a probabilistic approach for the automatic production of tree models with convincing 3D appearance and motion, and provides a generative model that creates multiple trees in 3D, given a single example model.
Shape from Silhouette Probability Maps: Reconstruction of Thin Objects in the Presence of Silhouette Extraction and Calibration Error
  • A. Tabb
  • Computer Science
    2013 IEEE Conference on Computer Vision and Pattern Recognition
  • 2013
An algorithm for an approximate solution using local minimum search for reconstructing the shape of thin, texture-less objects such as leafless trees when there is noise or deterministic error in the silhouette extraction step or there are small errors in camera calibration.
Nonrigid shape recovery by Gaussian process regression
A novel Gaussian process regression approach to the nonrigid shape recovery problem, which does not require to involve a predefined triangulated mesh model and is able to handle large deformations and outliers.
Learning to Boost Filamentary Structure Segmentation
  • Lin Gu, Li Cheng
  • Computer Science
    2015 IEEE International Conference on Computer Vision (ICCV)
  • 2015
This paper proposes an iterative two-step learning-based approach to boost the performance based on a base segmenter arbitrarily chosen from a number of existing segmenters, which has been empirically verified on specific filamentary structure segmentation tasks.
Principal Curves as Skeletons of Tubular Objects
This paper first introduces principal curves as a model for the underlying skeleton of axons and branches, then describes a recursive principal curve tracing (RPCT) method to extract this topology information from 3D microscopy imagery.
Graph cut based image segmentation with connectivity priors
This work forms several versions of the connectivity constraint and shows that the corresponding optimization problems are all NP-hard.