Soumya Ghosh
website  |   open in new window

My research interests lie in exploring the intersection of machine learning and computer vision. I am interested in identification, recognition and characterization of objects in cluttered natural scenes and in combining these algorithms to achieve higher level artificial intelligence tasks such as Image Understanding. I am inspired by biological vision processing systems and am also interested in developing machine vision systems mimicking them. On the machine learning front, I am interested in applying knowledge learnt from a previous task to a new domain ie,Transfer Learning. My work at CU focuses on developing solutions to these problems as applicable to mobile robotics.

 

Publications

Tomasz Stepinski, Soumya Ghosh, Ricardo Vilalta, Machine Learning for Automated Mapping of Planetary Surfaces, 19th Conference on Innovative Applications of Artificial Intelligence (IAAI), 2007
abstract | pdf

Soumya Ghosh, Tomasz Stepinski, Ricardo Vilalta, Automatic Mapping of Martian Landforms Using Segmentation-Based Classification, 38th Lunar and Planetary Science Conference (LPSC), 2007
abstract | pdf

Tomasz Stepinski, Soumya Ghosh, Ricardo Vilalta, Automatic Recognition of Landforms on Mars Using Terrain Segmentation and Classification, 9th International Conference on Discovery Science (DS), 2006
abstract | pdf

Soumya Ghosh, A Machine Learning Approach for Automated Geomorphic Map Generation, Masters Thesis, 2006
abstract | pdf

Soumya Ghosh, Frequency of Execution Metrics for GBT Observations, NRAO Technical Report, 2005
abstract | pdf