Daniel Heesch

I am interested in machine learning, information retrieval and, amongst other things, cycling. After a PhD in information retrieval and postdoctoral research in computer vision, I now spend much of my time providing research consultancy to technology companies, first and foremost Empora.com, the visual search engine for UK fashion.

Recent Activities

• Multi-bit quantisation for image similarity search (ISM 2009)
  This conference paper is a simple extension of the 1-bit quantisation of wavelet coefficients proposed by Jacobs, Finkelstein and Salesin in the mid-90s. We show that a quantiser with four partition regions improves retrieval results and prove a few theorems that help to find optimal partition sizes.
  
  
• Unsupervised Grabcut (ICIP 2009)
  We developed an unsupervised version of Grabcut in which the initialisation step is automated, and propose a new stopping criterion based on the KL divergence between background and foreground models.
  

View all publications

Education

2001 - 2005   PhD Computer Science, Imperial College London

2001 - 2005   BSc Mathematics, Open University

1997 - 2000   BA Biological Sciences, University of Oxford (St John's College)