Object recognition

Identification problem is
accomplished by using extracted
features
Definition:

– Segmentation/Figure-Ground Separation:
prerequisite or consequence?

– Labeling an object

– Extracting a parametric description as well

 Techniques

• Template matching

• Structural techniques
 Template matching
• Basic problem is to match the object with a stored
pattern stored (called model template) which is stored
during training procedure

• The features of object in a image are

-Area,
-Diameter,
-Aspect ratio etc compared with stored values

• Match is found allowing certain variations in comparison

process
 Models of Object Recognition
• Hierarchical Template Matching:
Image passed through layers of units with
progressively more complex features at
progressively less specific location. Hierarchical in
that features at one stage are built from features at
earlier stages.

• Several processing layers

- comprising simple (S)
- complex (C) cells.
• S-cells in one layer respond to conjunctions of C-
cells in previous layer.
• C-cells in one layer are excited by small
neighborhoods of S-cells.
 Transform & Match:
• First take care of rotation, translation, scale, etc. invariances.
• Then recognize based on standardized pixel representation
of objects.

• Template match: e.g.,

with an associative
memory based on
a Hopfield network.
 Structural techniques
• Relation between features of an object

• Complex objects are composed so simpler

pieces

• We can recognize a novel/unfamiliar object by

parsing it in terms of its component pieces, then
comparing the assemblage of pieces to those of
known objects.
 Recognition by Components
• GEONS: geometric elements of which all objects are
composed (cylinders, cones, etc). On the order of 30
different shapes.
• Skips 2 ½ D sketch: Geons are directly recognized from
edges, based on their nonaccidental properties (i.e., 3D
features that are usually preserved by the projective imaging
process).
 Recognition by Hypothesize and Test

• General idea
– Hypothesize object identity and pose
– Render object in camera
– Compare to image
• Issues
– where do the hypotheses come from?
– How do we compare to image (verification)?
Step 1: correspondence
 Features
They have to project like points
– Lines
– Conics
– Other fitted curves
– Regions (particularly the center of a region, etc.)
Step 2: Shape deformation and matching
 Pose consistency
• Strategy:
– Generate hypotheses using small numbers of
correspondences (e.g. triples of points for a
calibrated perspective camera, etc.)
– Backproject and verify

• Appropriate groups are “frame groups”

Models

Body Recognition
• How to find
landmarks to match
across two images?
• How to distinguish
one landmark from
another?
Face points filtered by known L-R, up-
down, and distance relationships

Points identified by surface curvature in neighborhood and filtered by

location relative to other salient points. These 3 points are then used for
iterative 3D alignment (ICP algorithm) of the scan to a 3D model face.
 Training vision system
• To program vision system with known
object
• System stores these objects in the form of
extracted feature values which can be
compared against the corresponding
feature