Where Data Meets Computer Science
Header

GPU

The local Gaussian distribution fitting (LGDF) energy model is a state-of-the-art method, capable of segmenting inhomogeneous objects with poorly defined boundaries, but it is computationally expensive. In our approach, we port the LGDF energy functional to the GPU, and introduce a novel set of interactive brush functions to segment challenging datasets where an active contour would not ordinarily capture the target object. Furthermore, we expose a smaller and more intuitive parameter space to the user, and enhance usability by including a built-in ray tracer to visualise the evolving 3D segmentation results in real time. Quantitive and qualititive validation is presented, demonstrating both a faithful port of the original algorithm and the practical efficacy of our interactive elements, for a wide variety real-world datasets.

Brain.

Brain.

Collaborators:

  • NVIDIA

Publications:

  • [2017,article] bibtex
    C. Willcocks, P. T. G. Jackson, C. J. Nelson, A. Nasrulloh, and B. Obara, "Interactive GPU Active Contours for Segmenting Inhomogeneous Objects," Journal of Real-Time Image Processing, 2017.
    @article{Willcocks2017,
      author = {Chris Willcocks and Philip T.G. Jackson and Carl J. Nelson and Amar Nasrulloh and Boguslaw Obara},
      title = {Interactive {GPU} Active Contours for Segmenting Inhomogeneous Objects},
      journal = {Journal of Real-Time Image Processing},
      year = {2017}
    }