【feature】Region detectors & descriptors

Affine Covariant Features

affine regions from
veiwpoint 1
Image 1
affine regions from
viewpoint 2
Image 2

This project focus on the problem of detecting affine invariant features in arbitrary images and on the performance evaluation of region detectors/descriptors.

  1. Region detector
    • Harris-Affine & Hessian Affine:
    • MSER ( Maximaly stable extremal regions):
    • IBR & EBR(Intensity extrema based detector & Edge based detector):
      • T.Tuytelaars and L. Van Gool, Matching widely separated views based on affine invariant regions . In IJCV 59(1):61-85, 2004. PDF
    • Salient regions:
    • All Detectors – Survey:
  1. Region descriptor
    • SIFT:
      •  D. Lowe, Distinctive image features from scale invariant keypoints. In IJCV 60(2):91-110, 2004. PDF
  2. Performance
  3. Datasets
    • images in PPM format
    • homographies between image pairs.
    • five different changes in imaging conditions:
      • viewpoint changes,
      • scale changes, image blur,
      • JPEG compression, and
      • illumination.
  • Blur

bikes
1000×700
6 images

  • Blur

trees
1000×700
6 images

  • Viewpoint
  • the viewpoint change test the camera varies from a fronto-parallel view to one with significant foreshortening at approximately 60 degrees to the camera

graffiti
800×640
6 images

  • Viewpoint

bricks
1000×700
6 images

  • Zoom+rotation:
    The scale changes by about a factor of four.

bark
765×512
6 images

  • Zoom+rotation

boat
800×640
6 images

  • Light

The light changes are introduced by varying the camera aperture.

cars
921×614
6 images

  • JPEG compression

The JPEG sequence is generated using a standard xv image browser with the image quality parameter varying from  40\% to 2\%

ubc
800×640
6 images

Updates

Binaries Date Comments
Harris & Hessian(also Windows)(1921206B) 8-6-2006 Scale & affine invariant feature detectors used in Mikolajczyk CVPR06 and CVPR08 for object class recognition. Efficient implementation of both, detectors and descriptors. Currently only sift descriptor was tested with the detectors but the other descriptors should work as well. Package contains a PCA basis for projections on fewer number of dimensions. Run with pca projection and take as many SIFT dimensions as you wish. Run without options for help. Includes windows executable which requires cygwin.

./extract_features -harhes -i img.png -sift -pca harhessift.basis
Scale Saliency 8-9-2006 Some problems were reported for the Salient region detector. Please, try the new versions from the author’s homepage.
Detectors & Descriptors 12-6-2007 Same as Harris and Hessian above but more parameters are made available for setting, a few bugs fixed etc. It’s suitable for extracting features for recognition in terms of feature number and type.
SURF detector 10-7-2007 Speeded Up Robust Features – fast implementation of SIFT using integral images.
FAST corner detector 10-7-2007 Features from Accelerate

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

生活在西班牙

自己动手丰衣足食

BlueAsteroid

Just another WordPress.com site

Jing's Blog

Just another WordPress.com site

Start from here......

我的心情魔方

天才遠私廚

希望能做一個分享各種資訊的好地方

语义噪声

西瓜大丸子汤的博客

笑对人生,傲立寰宇

Just another WordPress.com site

Where On Earth Is Waldo?

A Project By Melanie Coles

the Serious Computer Vision Blog

A blog about computer vision and serious stuff

Cauthy's Blog

paper review...

Cornell Computer Vision Seminar Blog

Blog for CS 7670 - Special Topics in Computer Vision

datarazzi

Life through nerd-colored glasses

Luciana Haill

Brainwaves Augmenting Consciousness

槑烎

1,2,∞

Dr Paul Tennent

and the university of nottingham

turn off the lights, please

A bunch of random, thinned and stateless thoughts around the Web

%d bloggers like this: