home   |   research   |   teaching   |   publications   |   biosketch   |   CV

HMA Feature-Matching Toolbox

ASTRA Robotics Lab - University of Texas at Arlington

Gustavo Puerto and Dr. Gian Luca Mariottini

1  Project Overview

The ability to find image similarities (feature matching) between laparoscopic views is essential in many robotic-assisted Minimally-Invasive Surgery (MIS) applications. Differently from feature tracking, feature-matching methods do not make any restrictive assumption about the sequential nature of the two images or about the organ motion. As a consequence, they can be used fundamentally to recover those tracked features that were lost due to a prolonged occlusion, a sudden endoscopic-camera retraction, or a strong illumination change. We present here the Feature-Matching toolbox which includes several recent algorithms, and our novel Hierarchical Multi-Affine (HMA) feature-matching algorithm, which improves over existing methods because of the larger number of image correspondences, the increased speed, and the higher accuracy and robustness.

Related Publications

Please cite the following publication when using our toolbox:
G.A. Puerto and G.L. Mariottini. "A Fast and Accurate Feature-Matching Algorithm for Minimally Invasive Endoscopic Images". IEEE Transactions on Medical Imaging, 2013 (in press)

2  Feature-Matching Toolbox

The HMA Feature-Matching Toolbox includes four recent algorithms [1,2,3,4] as well as the above IEEE TMI journal-paper version. The toolbox includes a demo, to show how to run the different algorithms.
The Feature-Matching Toolbox requires to download a pair of libraries to run the implementation with all the options. These libraries can be download from the links in the download section. Consult README file in the Feature-Matching Toolbox for more details.

3  Hierarchical Multi-Affine (HMA) toolbox

The HMA Toolbox includes a demo, to show how to run the different algorithms.
The HMA Toolbox requires to one extra library (VLFeat). This library can be downloaded from the links in the download section. Consult the README file in the HMA Toolbox for more details.

4  Download

Download: Feature-Matching Toolbox Version 1.0 (Last update at June 17 2012)
HMA Tolbox Version 1.1 (Last update on Jan.26th, 2013)
In-Lab Databases: SIFT - SURF - ASIFT
Link: VLFeat Library Version 0.9.14
Link: OpenSurf Last update at 06 Sep 2010


D.G. Lowe. Distinctive image features from scale-invariant keypoints. Int. J. Comp. Vis., 60(2):91-110, 2004.
M. Cho, J. Lee, and K.M. Lee. Feature correspondence and deformable object matching via agglomerative correspondence clustering. In Proc. 9th Int. Conf. Comp. Vis., pages 1280-1287, Sept. 2009.
A. Del Bimbo, F. Franco, and F. Pernici. Local shape estimation from a single keypoint. In Proc. Comp. Vis. Patt. Rec. Workshops, pages 23-28, June 2010.
G. Puerto-Souza, M. Adibi, J. A. Cadeddu, and G. Mariottini. Adaptive multi-affine (ama) feature-matching algorithm and its application to minimally-invasive surgery images. In Proc. IEEE/RSJ Int. Conf. Intel. Rob. Syst, pages 2371 -2376, Sept. 2011.

File translated from TEX by TTH, version 3.85.
On 26 Jan 2013, 15:29.