Lambert Schomaker
NICI / Nijmegen University
The Netherlands
hwr.nici.kun.nl
QBIC (IBM)
... but what was the user's intention with the query (top left photo = query)?
... cumbersome, feature selection/weighing requires knowledge ...
FourEyes (MIT Medialab)
(b): pen-based image annotation and queries
Acronyms galore
(b): pen-based image annotation and queries
Existing Methods
VisualSEEk
... relation between texture and object-related content unclear,
arbitary segmentation creates ambiguity...
(b): pen-based image annotation and queries
Query types
Query
Matched with:
Matching algorithm
A
keywords
manually provided textual image annotations
free text and information-retrieval (IR) methods
B
keywords
textual and contextual information in the image neighbourhood
free text and IR methods
C
exemplar image
image bitmap
template matching or feature-based
D
rectangular sub image
image bitmap
template matching or feature-based
E
layout structure
image bitmap
texture and color segmentation
F
object outline
image bitmap, contours
feature-based
G
object sketch
image bitmap
feature-based
(excluded from this table are: point & click navigation in systematically organized image bases)
(b): pen-based image annotation and queries
Problems
queries in a user community?
(b): pen-based image annotation and queries
Ergonomic, Cognitive & Perceptual aspects
(b): pen-based image annotation and queries
How to realize such object-based image search?
(b): pen-based image annotation and queries
Design considerations
(b): pen-based image annotation and queries
(b): pen-based image annotation and queries
Example: 'bodyworks' shape of motor bicycles
(note the distribution of points of high curvature along the outline)
Simple 1-NN matching will be used for all feature categories.
Outline matching: best starting point and clockwise/counterclockwise search.
Given a precision proportion q for the result of a particular query,
there will be a number of x = n q correct items in the hitlist.
For a meaningful result we want q >> p(X = x).
i.e., Finding 1 hit in list of 16 is not so unlikely: p = 0.38
A query to find an engine
(b): pen-based image annotation and queries
Annotation
(b): pen-based image annotation and queries
Algorithm: a few matching possibilities
(b): pen-based image annotation and queries
System architecture
(b): pen-based image annotation and queries
Performance measurement aspects
(b): pen-based image annotation and queries
Hit list: accident or meaningful?
(b): pen-based image annotation and queries
Hit list: accident or meaningful?
Assume a collection of N items where r(ight) of these items are of
one type (type A) and the remaining N-r are of another type (type B).
Wanted is the probability of obtaining exactly X items of type
A in a subset of n elements randomly drawn from the total of N items.
Then X is distributed according to the Hypergeometric Distribution:
Example (given N=750 images in total
r=50 instances in target class,
n=16 images in hit list):
P(X=0) = 0.33
P(X=1) = 0.38
P(X=2) = 0.21
(b): pen-based image annotation and queries
Test the concept: Can the users do it?
Number of subjects producing an outline: 33, number of photographs: 10.
Photographs: brain, Buddha christmas tree, monster truck,
jukebox,locomotive, motor cycle, mushroom cloud, pistol.
Results kindly provided by Arie Baris.
(b): pen-based image annotation and queries
Test the concept: Can the users do it?
(b): pen-based image annotation and queries
Hierarchical clustering on outlines
(b): pen-based image annotation and queries
(b) Conclusion
translation, scale, rotation and mirror invariance must be solved!
This is easy in outline-outline matching.
(b): pen-based image annotation and queries
(b) Conclusion (continued)
Tutorial "new pen-based applications" ICDAR'99 Bangalore. Copyright 1999 L. Schomaker