I am currently a PhD student within the SketchX Research Lab and the Computer Vision Group in the School of Electronic Engineering and Computer Science at Queen Mary, University of London, jointly supervised by Dr. Yi-Zhe Song, Dr. Tao Xiang, and Dr. Timothy Hospedales .

Before joining the Computer Vision Group in QMUL I received my B.Sc. in Electronics and Information Engineering from Dalian University of Technology. Then after 3 years research in image and speech processing, I received my M.Sc in AAU Lab at Dalian University of Technology, jointly supervised by Prof. Xiaohong Ma and Prof. Shuxue Ding.

Currently I am interested in sketch-based image retrieval using Deep Learning. I am now developing framework learning cross-domain representation for sketch and photo modality, and then make fine-grained retrieval based on the learned representation. Besides, fine-grained attributes are also exploited in the multi-task learning.

Education Background

  • B.Sc. in Electronic and Information Engineering, Dalian University of Technology, 2012
  • M.Sc. in Signal and Information Processing, Dalian University of Technology, 2015


  1. Deep Multi-task Attribute-based Ranking for Fine-grained Sketch-based Image Retrieval PDF Dataset
    Jifei Song, Yi-Zhe Song, Tao Xiang, Timothy Hospedales and Xiang Ruan
    BMVC, 2016.


  1. Fine-Grained Image Retrieval: the Text/Sketch Input Dilemma PDF Dataset
    Jifei Song, Yi-Zhe Song, Tao Xiang and Timothy Hospedales
    BMVC, 2017.
  2. Deep Spatial-Semantic Attention for Fine-Grained Sketch-Based Image Retrieval PDF Dataset Project
    Jifei Song*, Qian Yu*, Yi-Zhe Song, Tao Xiang and Timothy Hospedales
    ICCV, 2017.


  1. Learning to Sketch with Shortcut Cycle Consistency PDF arXiv Demo
    Jifei Song, Kaiyue Pang, Yi-Zhe Song, Tao Xiang and Timothy Hospedales
    CVPR, 2018.

Design the fine-grained SBIR online demo