Dress.com: A CBIR-Based Digital Paper Doll System for Real-Life Dressing

Chueh-Min Cheng Chun-Wei Liu

National Taiwan University



We present the “Dress.com” system, a content-based image retrieval system which helps user to search their virtual wardrobes by photos. Once set up, the system allows users to virtually change their clothing by simply take a picture of target clothing item. A long-term goal of the system is to help users to find appropriate clothing collocations.


Download the pdf file Paper:
7-page PDF


  1. “Comparison of texture features based on Gabor filters.” [Link]
    S.E. Grigorescu, Nicolai Petkov, and P. Kruizinga.
    IEEE Transactions on Image processing, 2002
  2. “GABORCONVOLVE - function for convolving image with log-Garbor filters.” [Link]
    Peter Kovesi.
  3. “Lazy Snapping.” [Link]
    Yin Li, Jian Sun, Chi-Keung Tang, and Harry Shum.
    SIGGRAPH 2004
  4. “Face description with local binary patterns: Application to face recognition.” [Link]
    Timo Ahonen, Abdenour Hadid, and Matti Pietikäinen.
    PAMI 2006