Proposal of CAPTCHA using three dimensional objects

Authors:
Kenta Kunitani;Ryuya Uda
Affiliations:
Tokyo University of Technology, Hachioji City, Japan;Tokyo University of Technology, Hachioji City, Tokyo, Japan
Venue:
Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication
Year:
2013

Citing 3
Cited 0

Recognizing objects in adversarial clutter: breaking a visual captcha

CVPR'03 Proceedings of the 2003 IEEE computer society conference on Computer vision and pattern recognition
Building segmentation based human-friendly human interaction proofs (HIPs)

HIP'05 Proceedings of the Second international conference on Human Interactive Proofs
Proposal of Movie CAPTCHA Method Using Amodal Completion

SAINT '12 Proceedings of the 2012 IEEE/IPSJ 12th International Symposium on Applications and the Internet

Quantified Score

Hi-index	0.00

Visualization

Abstract

Over the last few years, unfair use of web services by robot software has grown a quite big problem. In order to prevent the robot software from obtaining new accounts of web services, CAPTCHA is popularly used. CAPTCHA is a method which confirms whether a user is a human being or robot software. The most popular CAPTCHA is a method to read distorted letters. However, it is analyzed by computer software according to developing OCR technology. As a countermeasure against the OCR technology, strongly distorted letters for CAPTCHA are proposed. However, it is hard for human beings to read them. Moreover, there appears a new proposal in which movie is used for CAPTCHA with amodal completion which is one of the human abilities for sight. However, if performance of computer is developed, that proposal might become out of use. Therefore, in our proposal, amodal completion is applied to three dimensional objects in order to increase the calculation cost for a computer up to the level of practical use.