TR2011-037

CrossTrack: Robust 3D Tracking from Two Cross-Sectional Views


    •  Hussein, M., Porikli, F., Li, R., Arsian, S., "CrossTrack: Robust 3D Tracking from Two Cross-Sectional Views", IEEE Conference on Computer Vision and Pattern Recognition (CVPR), DOI: 10.1109/​CVPR.2011.5995429, June 2011, pp. 1041-1048.
      BibTeX TR2011-037 PDF
      • @inproceedings{Hussein2011jun,
      • author = {Hussein, M. and Porikli, F. and Li, R. and Arsian, S.},
      • title = {CrossTrack: Robust 3D Tracking from Two Cross-Sectional Views},
      • booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
      • year = 2011,
      • pages = {1041--1048},
      • month = jun,
      • doi = {10.1109/CVPR.2011.5995429},
      • url = {https://www.merl.com/publications/TR2011-037}
      • }
  • Research Area:

    Computer Vision

Abstract:

One of the challenges in radiotherapy of moving tumors is to determine the location of the tumor accurately. Existing solutions to the problem are either invasive or inaccurate. We introduce a non-invasive solution to the problem by tracking the tumor in 3D using bi-plane ultrasound image sequences. We present CrossTrack, a novel tracking algorithm in this framework. We pose the problem as recursive inference of 3D location and tumor boundary segmentation in the two ultrasound views using the tumor 3D model as a prior. For the segmentation task, a robust graph-based approach is deployed as follows: First, robust segmentation priors are obtained through the tumor 3D model. Second, a unified graph combining information across time and multiple views is constructed with a robust weighting function. For the tracking task, an effective mechanism for recovery from respiration-induced occlusion is introduced. Our experiments show the robustness of CrossTrack in handling challenging tumor shapes and disappearance scenarios, with sub-voxel accuracy, and almost 100% precision and recall, significantly outperforming baseline solutions

 

  • Related News & Events

    •  NEWS    CVPR 2011: 6 publications by Yuichi Taguchi, Srikumar Ramalingam, Amit K. Agrawal and C. Oncel Tuzel
      Date: June 21, 2011
      Where: IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
      Research Area: Computer Vision
      Brief
      • The papers "Entropy Rate Superpixel Segmentation" by Liu, M.-Y., Tuzel, O., Ramalingam, S. and Chellappa, R., "Structured Light 3D Scanning in the Presence of Global Illumination" by Gupta, M., Agrawal, A., Veeraraghavan, A. and Narasimhan, S., "CrossTrack: Robust 3D Tracking from Two Cross-Sectional Views" by Hussein, M., Porikli, F., Li, R. and Arsian, S., "P2C2: Programmable Pixel Compressive Camera for High Speed Imaging" by Reddy, D., Veeraraghavan, A. and Chellappa, R., "Beyond Alhazen's Problem: Analytical Projection Model for Non-Central Catadioptric Cameras with Quadric Mirrors" by Agrawal, A., Taguchi, Y. and Ramalingam, S. and "The Light-Path Less Traveled" by Ramalingam, S., Bouaziz, S., Sturm, P. and Torr, P. were presented at the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
    •