MindFinder: Finding Images by Sketching

Project Descriptions

Sketch-based image search is a well-known problem difficult to solve. Little progress has been made in the past decade in developing a large-scale, practical, sketch-based search engine. We have revisited this problem and developed a scalable solution. Based on this solution, we have built a system called MindFinder that indexes more than 2 million web images, enabling efficient, sketch-based image retrieval. MindFinder enables users to sketch major curves of a target image as a query. Based on these sketches, relevant images can be returned in real time. The system also supports the addition of semantic tags and colors to enhance the relevance of search results.

 

Videos

 

Softwares

Sketch Match: On 2011-08-17, the app, Sketch Match, was released to WP7 marketplace! It is powered by the MindFinder technology (sketch-based image search), with a well-refined sketch matching game.

 

Online Systems

Natural Image Search Engine

(Microsoft corpnet available only.)

 

 

 

 Clipart Image Search Engine

 (Microsoft corpnet available only.) 

 

 

 

Publications

  • The paper "Sketch2Tag: Automatic Hand-Drawn Sketch Recognition" (Demo) was accepted by ACM Multimedia 2012.

  • The paper "Query-Adaptive Shape Topic Mining for Hand-Drawn Sketch Recognition" was accepted by ACM Multimedia 2012.

  • The paper "Free Hand-Drawn Sketch Segmentation" was accepted by ECCV 2012.

  • Xianming Liu, Changhu Wang, Hongxun Yao, Lei Zhang, The Scale of Edges, in CVPR 2012.

 People

Changhu Wang, Lei Zhang, Bruce Yang, Xin Zou, Zhiwei Li, Yang Cao, Xianming Liu, Hai Wang*

 *past interns 

 

 Contact

Changhu Wang

Email: chw AT microsoft DOT com

Share
Share this page on Facebook
Share this page on Twitter
Share this page on LinkedIn
E-mail this page
RSS feeds