Short Biography
I am now at the Laboratoire ERIC, Université Lumière Lyon 2, working as a Post-doctoral researcher. I am conducting my research activities to develop novel clustering method for dynamic time series data.
I recently completed my Ph.D. in "image, vision and signal" at the Laboratoire Hubert Curien UMR CNRS 5516, Université Jean Monnet Saint Etienne, funded by ARC 6 - Région Rhône-Alpes. I was supervised by Prof. Olivier ALATA and Prof. Alain TREMEAU. I have conducted research in the area of Machine Learning and Computer Vision. More specifically, I worked on statistical mixture models, scene segmentation and three dimensional modeling. Besides, I am also interested to conduct research on Natural Language Processing, Document Analysis and Recognition and Spectral Image Processing.
I have completed my Master’s from the Erasmus Mundus Masters Degree program entitled “Color in Informatics and Media Technology (CIMET)”. In my Masters I have studied at the University of Eastern Finland, Finland and at the University of Granada, Spain.
I worked as a researcher at the Center for Research on Bangla Language Processing (CRBLP) from September 2006 to August 2009. My major responsibility was to develop an Optical Character Recognizer for Bangla Script. Till date I have conducted research on Natural Language Processing, Pattern Recognition, Document Image Analysis and Recognition and Speech Recognition. Moreover, I am interested to conduct research on Natural Language Understanding and Artificial Intelligence.
I completed my B.Sc. degree on Computer Science and Engineering from BRAC University at 2006 with CGPA 3.8/4.0. In my undergraduate thesis I developed “Real Time Traveler Information System” under the supervision of Professor Mumit Khan. I conducted research on “feature extraction of Bangla characters” and “development of an HMM based Bangla OCR” at CRBLP while I was still an undergraduate student.
I started my professional career at CRBLP. Till date I developed the first open source Bangla OCR (2006), limited vocabulary Bangla isolated and continuous Automatic Speech Recognizer (2007), Bangla Character Segmenter (2008), Tesseract Based Bangla OCR (2008). Beside these, I conducted my research with two Google sponsored open source OCR projects Tesseract and OCROpus. My key responsibility was to integrate Bangla script recognition support into these projects. As part of the research work I released Tesseract training data for Bangla scripts and a tool to convert text to image. Beside these research activities, I also worked on emotional Bangla voice recognition. Aside from my research and development, I was a laboratory instructor of the undergraduate course Compiler Design from fall 2007 to summer 2008. Additionally I co-supervised several student thesis projects with Professor Khan as well as supervised the interns at CRBLP.
Till date ten of my research papers have been accepted and published in multiple international conferences.