SLIDE 1
Image Features: Detection, Description, and Matching and their - - PowerPoint PPT Presentation
Image Features: Detection, Description, and Matching and their - - PowerPoint PPT Presentation
Image Features: Detection, Description, and Matching and their Applications Image Representation: Global Versus Local Features Features/ keypoints/ interset points are interesting locations in the image. Features serve to give a new
SLIDE 2
SLIDE 3
Features/ keypoints/ interset points are interesting locations in the image. Features serve to give a new compact representation of an image. Features can describe objects. Types of image features: Edges, Corners, blobs, small image patches
SLIDE 4
Global Features Local Features
Image is represented by one multidimensional feature vector, describing the information in the whole image. Imagemayhavehundredsof localfeatures. Measure various aspects of the image such as color, texture or shape. Measure local structures that are more distinctive and stable than other structures. Much faster and compact while easy to compute and generally require small amounts of memory. Local features have superior performance. Not invariant to significant transformations and sensitive to clutter and occlusion. Robust against occlusion and background clutter.
SLIDE 5
Features
Corner Region Edge
SLIDE 6
Features Detection:
Detection: identify the interest points Feature detection = how to find some interesting points (features) in the image.(Ex: find a corner, find a region and so on...)
SLIDE 7
Features Detection:
Detectors Single-scale Harris FAST SUSAN Multi-scale Harris- Laplace Hessian- Laplace LOG DOG Affine- invariant Harris- Affine Hessian
- Affine
SLIDE 8
SLIDE 9
Features description:
Description: Extract feature vector surrounding each interest point. Feature description = how to represent the interesting points we found to compare them with other interesting points (features) in the image.
SLIDE 10
Features description
Descriptors Gradient-based (Floating – point) SIFT SURF GLOH Binary-based (Binary string) BRIEF ORB BRISK FREAK
SLIDE 11
Features Matching:
Matching: match/compare the extracted descriptor from the query image to those in the database. Feature Matching= find the distance between two
- r more descriptors/ finding the similarity.
SLIDE 12
Features Matching:
Given two such patches, how can we determine their similarity? We can measure the pixel to pixel similarity by measuring their Euclidean distance, but that measure is very sensitive to noise, rotation, translation and illumination changes. In most applications we would like to be robust to such changes.
SLIDE 13
SLIDE 14
SLIDE 15
Applications
1.Object Recognition
- 2. Camera Calibration
- 3. Image Retrieval
- 4. Image Registration
SLIDE 16
Binary Descriptors
Binary descriptors are composed of three parts:
- A sampling pattern: where to sample points in
the region around the feature/keypoint.
- Orientation compensation: some mechanism to
measure the orientation of the keypoint and rotate it to compensate for rotation changes.
- Sampling pairs: which pairs to compare when
building the final descriptor.
SLIDE 17
Why binary descriptors?
- Pixel comparisons are much faster than
gradients.
- Matching binary representations is done using
hamming distance which is much faster to compute than any distance metric.
- Whereas a single gradient based descriptor
commonly require 64 or 128 floating point values to store, a single binary descriptors requires only 512 bits – 4 to 8 times less.
SLIDE 18
Challenges
- Change of scale
- Change of orientation (rotation)
- Change of viewpoint (affine, projective
transformations)
- Change of illumination
- Noise, Clutter, and occlusion
- Repetitive patterns (windows, street marks, etc.)
Thus, we need robust similarity measures, detector, descriptors, and matchers
SLIDE 19
Ideas
- Studying the effect of different sampling
patterns on descriptors’ performance.
- Investigating the effect of different similarity
metrics on the matching performance.
- Adaptation of binary descriptors for
computational resources.
- Enhancing existing descriptors’ robustness
against geometric and photometric transformations.
- Proposing our own descriptor and using it in
applications.
SLIDE 20
Thank you
SLIDE 21
A schematic representation of the SIFT descriptor for a 16 × 16 pixel patch and a 4 × 4 descriptor array
SLIDE 22