当前位置：文档库 › 11_ObjectRecognition

11_ObjectRecognition

Given a database of objects and an

CS 461, Copyright G.D. Hager

Usually, choose k such that ,k / ,1 < where is small (e.g. .05).Call the resulting matrix E (for eigenvalue projection).

I 0i } for each image i of object o

12/3/2003CS 461, Copyright G.D. Hager

An example: surfaces of first 3

coefficients

12/3/2003CS 461, Copyright G.D. Hager

g C

there are faster techniques (e.g. k-d trees) for doing this Return O as the identification of the object

Courtesy Shree Nayar, Columbia U. 12/3/2003CS 461, Copyright G.D. Hager

CS 461, Copyright G.D. Hager

(liberally borrowed from Embick and Marcus)

Convexity Labeling Conventions

A line labeled plus (+) indicates that the corresponding edge is convex ;

A line labeled minus (-) indicates that the corresponding edge is concave;

arrow indicates an occluding edge. To its right is the body

for which the arrow line provides an edge. On its left is space.

CS 461, Copyright G.D. Hager

(liberally borrowed from Embick and Marcus)

12/3/2003CS 461, Copyright G.D. Hager

must assign the same line label to any given line.

(liberally borrowed from Embick and Marcus

12/3/2003CS 461, Copyright G.D. Hager

An Example of Edge Consistency

?Consider an arrow junction with an L junction to the right:

?A1 and either L1 or L6 are compatible since they both

associate the same kind of arrow with the line.

?A1 and L2 are incompatible ,

since the arrows are pointed in the opposing directions,

?Similarly, A1 and L3 are incompatible.

(liberally borrowed from Embick and Marcus)

CS 461, Copyright G.D. Hager

(liberally borrowed from Embick and Marcus)

3/2/2 /20/203/2

03/

03/2/2 /20/203/2

/2/23/

3/23/

2/20

3/20/2

3/20