Image annotation combines machine vision and crowd-sourcing

Aug. 1, 2012
A PhD student at the University of California, San Diego (USCD; San Diego, CA, USA) has received an award to investigate the viability of an image annotation system that uses a combination of machine vision and crowd-sourcing.

A PhD student at the University of California, San Diego (USCD; San Diego, CA, USA) has received a Small Business Innovation Research award from the National Science Foundation (NSF) to investigate the viability of an image annotation system that uses a combination of machine vision and crowd-sourcing.

Dr. Boris Babenko believes that a system capable of annotating a large body of images quickly, accurately and inexpensively would be a valuable capability in medical diagnosis, neuroscience and urban planning, but the broader impact of the project could prove useful in more applications, as the rise in imaging data increases.

Research at Caltech (Pasadena, CA, USA) and UC San Diego suggests that it is possible to combine the complementary strengths of human annotators and machines into a hybrid system that is flexible, accurate, fast and inexpensive. Now, Babenko aims to build a prototype system to test the hypothesis.

By combining computer vision and machine learning automation with humans, both experts and non-expert annotators, Babenko believes that his system will be configurable enough such that it could address virtually any image analysis challenge.

Recently, Babenko and his colleagues founded a new company called Anchovi Labs (San Diego, CA, USA) which will offer image annotation and analysis services for scientific and industrial applications.

-- Dave Wilson, Senior Editor, Vision Systems Design

Voice Your Opinion

To join the conversation, and become an exclusive member of Vision Systems Design, create an account today!