Local Depth SIFT and Scale Invariant Spin Image local features for 3D meshes

a toolbox to compute Local Depth SIFT and Scale Invariant Spin Image local features for 3D meshes.
3,1K download
Aggiornato 17 dic 2012

Visualizza la licenza

descriptor matlab toolbox - a toolbox containing my work on local features for meshes: MeshScaleDoG local features detector, the Scale Invariant Spin Image descriptor and the Local Depth SIFT descriptor.

if you use this code please cite: Tal Darom and Yosi Keller, “Scale Invariant Features for 3D Mesh Models”. IEEE Transactions on Image Processing 21(5): 2758-2769 (2012).

The examples include models taken from:
Doug M. Boyer, Yaron Lipman, Elizabeth St. Clair, Jesus Puente, Biren A. Patel, Thomas Funkhouser, Jukka Jernvall, and Ingrid Daubechies. 'Algorithms to Automatically Quantify the Geometric Similarity of Anatomical Surfaces'. PNAS, vol. 108 no. 45, November 8, 2011

before using the toolbox please run compile_mex_files.m to compile all mex files.

main functions:

dog( vertex, faces) - runs the MeshScaleDoG detector on the mesh.

create_mesh_sift_features(vertex, faces) - runs the MeshScaleDoG detector on the mesh and extract the LD-SIFT descriptors for these points. It needs the VLFeat toolbox installed.

create_mesh_spin_features( vertex, faces) - runs the MeshScaleDoG detector on the mesh and extract the SISI descriptors for these points.

find_candidates_example - loads 2 mesh models, calculates descriptors for both and for each feature point of the first model finds matching candidates on the second based on descriptor distance.

This toolbox contains some code taken from "Toolbox Graph" by Gabriel Peyre and from "Patch Software Render" by Dirk-Jan Kroon.

Cita come

Tal Darom (2024). Local Depth SIFT and Scale Invariant Spin Image local features for 3D meshes (https://www.mathworks.com/matlabcentral/fileexchange/36316-local-depth-sift-and-scale-invariant-spin-image-local-features-for-3d-meshes), MATLAB Central File Exchange. Recuperato .

Compatibilità della release di MATLAB
Creato con R2010a
Compatibile con qualsiasi release
Compatibilità della piattaforma
Windows macOS Linux
Riconoscimenti

Ispirato da: Toolbox Graph, Patch Software Render

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!
Versione Pubblicato Note della release
1.2.0.0

fixed a small bug suggested by one of the comments. Added a usage example.

1.1.0.0

added the published paper details.

1.0.0.0