MARQS: RETRIEVING SKETCHES USING DOMAIN- AND STYLE-INDEPENDENT
FEATURES LEARNED FROM A SINGLE EXAMPLE USING A DUAL-CLASSIFIER
Brandon Paulson, Tracy Hammond
Paulson's system was designed with three goals: 1)recognition of a sketch regardless of drawing style and invariantly with scale and rotation, 2)beginning recognition with only a single example, and 3)learning over time. This is done by using a small number of global features for a single and a multiple example classifier. Sketches that are correctly recognized are added to the training examples, which initially consist of only a single drawing. To recognize a sketch, an input it first rotated so that the major axis is horizontal, eliminating variation due to rotation. Then four features are computed and fed to the classifiers. These features are: aspect ratio, pixel density, average curvature, and number of perceived corners. The single classifier is a simple nearest neighbor classifier, and the multi-example uses a linear classifier like that of Rubine. To test the system, 150 sketches were generated, representing 15 different symbols. The first sketch of each symbol was used as the training example. The search system relied on the linear classifer 73% of the time and the correct symbol was identified in the top 2 results on average.
The big issue with this system that I see it how well it would scale. The number of symbols used in the experiments, though possibly suitable for a media library search, seems rather small for the e-journal example presented in the associated slides. Not only the slow down and accuracy loss associated with adding a new sketch seem to be a costly downside of the system, but the fact that local differences and rotation are ignored may cause symbols that the user perceives as different to be recognized as the same symbol. Some symbols depend on orientation for meaning (<>) as well as local differences (the happy and frowny faces example).
skip to main |
skip to sidebar
Summary and discussion of assigned readings for the Fall 20007 Sketch Recognition class
About Me
The Required Stuff
Year: PhD 3rd year
Email: p r i s c u s 2 7 @ gmail.com
Academic Interests: Neural Networks, Human Cognition, Manifold Learning
Relevant Experience: Classes - Machine Learning, Pattern Classification
Why are you taking this class?
Not only does the subject sound interesting, it also seems like a technology that will be very relevant to the future use of computers. Also, it sounded like a cool class to gain some knowledge in the area of human-computer interaction in a hands-on way.
What do you hope to gain?
A better understanding of how humans use handwritten/drawn information to communicate and also how we understand this information. How people recognize what is depicted.
5 to 10 years from now?
Not really sure. Hopefully I'll have graduated. I'd like to go into a teaching career, preferably at the university level, though teaching at a lower level of education and being able to provide an introduction to computer science at the high school or younger level would also be interesting.
Nonacademic interests:
Swing (and other forms of) dancing, Computer and console gaming, Sci-fi and fantasy literature.
Fun Story:
Since this has recently been told to everyone I know, I'll tell it. When I was about 5 or 6 years old, we were heading out to the car to go to church. I was dressed to the nines, new shoes, new slacks, a button up shirt, hair slicked back. My mom looks over at me and says, "Well Paul, you sure look sharp today." And I instantly burst into tears. Because as we all know, one of the very first things you learn is that sharp is very, very bad and you should never touch sharp things.
Email: p r i s c u s 2 7 @ gmail.com
Academic Interests: Neural Networks, Human Cognition, Manifold Learning
Relevant Experience: Classes - Machine Learning, Pattern Classification
Why are you taking this class?
Not only does the subject sound interesting, it also seems like a technology that will be very relevant to the future use of computers. Also, it sounded like a cool class to gain some knowledge in the area of human-computer interaction in a hands-on way.
What do you hope to gain?
A better understanding of how humans use handwritten/drawn information to communicate and also how we understand this information. How people recognize what is depicted.
5 to 10 years from now?
Not really sure. Hopefully I'll have graduated. I'd like to go into a teaching career, preferably at the university level, though teaching at a lower level of education and being able to provide an introduction to computer science at the high school or younger level would also be interesting.
Nonacademic interests:
Swing (and other forms of) dancing, Computer and console gaming, Sci-fi and fantasy literature.
Fun Story:
Since this has recently been told to everyone I know, I'll tell it. When I was about 5 or 6 years old, we were heading out to the car to go to church. I was dressed to the nines, new shoes, new slacks, a button up shirt, hair slicked back. My mom looks over at me and says, "Well Paul, you sure look sharp today." And I instantly burst into tears. Because as we all know, one of the very first things you learn is that sharp is very, very bad and you should never touch sharp things.
No comments:
Post a Comment