Research


My research interests include pattern recognition, computer vision and to some extent networking and systems.
But I'll enjoy just about anything related to computer-automation :)

Reviewer

Data Mining and Knowledge Discovery

List of Publications

  1. [PDF] Ohil K. Manyam, Neeraj Kumar, Peter Belhumeur & David Kriegman, "Two faces are better than one: Face recognition in group photographs", In proceedings of the IEEE International Joint Conference on Biometrics, Oct. 2011.
  2. [PDF] L.M. Patnaik & Ohil K. Manyam, "Epileptic EEG detection using neural-networks and post-classification", Computer Methods and Programs in Biomedicine, Issue 91/2, Aug. 2008, pages 100-109.
  3. [PDF] Amitav Das, Ohil K. Manyam & Makarand Tapaswi, "Multi-feature audio-visual person recognition", In proceedings of the IEEE International Workshop on Machine Learning and Signal Processing (MLSP), Cancun, Mexico, Oct 16-19, 2008, pages 227-232.
  4. [PDF] Amitav Das, Ohil K. Manyam & Makarand Tapaswi, "Audio-visual person authentication with multiple "Visualized-Speech Features" and multiple face profiles", In proceedings of the sixth IEEE Indian Conference on Computer Vision, Graphics & Image Processing (ICVGIP), Bhubaneswar, India, Dec 16-19, 2008, pages 39-46.
  5. [PDF] Amitav Das, Ohil K. Manyam, Makarand Tapaswi & Veeresh Taranalli, "Multilingual spoken-password based user authentication in emerging economies using cellular phone networks", In proceedings of the IEEE Workshop on Spoken Language Technology (SLT), Goa, India, Dec 15-18, 2008, pages 5-8.

Internships/Fellowships

  • Apple Inc. Cupertino, Summer 2010
    Worked in the Frameworks Group with members from the Hand-Writing team as a Software Engineering Intern. Involved in researching and implementing a confidential optical character recognition framework capable of achieving very high accuracy. Experimented and employed various algorithms from computer vision and machine learning. Built a Cocoa frontend in Objective-C for demonstration purposes. Presented my work to Craig Federighi, VP Mac OS Engineering. Was among five software interns across Apple in 2010 selected to present to Bertrand Serlet, Senior VP of Software Engineering.
  • Microsoft Research India, Summer 2009
    Worked in the Advanced Development and Prototyping Group as a Research Developer on contract. Designed and developed a web-based Wake-On-LAN & Remote Management Platform (WARP) for remote power management of computers on the Microsoft Corportate Network. Features included remote peer-to-peer wakeup, remote sleep, hibernate, and shutdown of a desktop. These asynchronous machine-state transitions were based on events published from user interfaces such as the Web and e-mail. Automatic client software upgrade, centralized system control, and auditing were also built into the system.
    Also involved in Baaz - a system for the automatic detection of accessibility and security anomalies in network Access Control Lists. This was in collaboration with the Mobility, Networks, and Systems group at MSRI.
  • Microsoft Research India, June-November, 2008
    Worked under Dr. Amitav Das of Microsoft Research India, as a Research Consultant, and continued to work deeper on the topics that were a part of the earlier internship. Developed improvements to the existing speaker and face identification algorithms. Was also involved in collecting a customized database for mobile-phone based speaker recognition. One research paper was accepted into an international IEEE conference. Presented a paper from my internship at the Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP) in Bhubaneswar, India (15th to 19th December 2008). Also demonstrated a face identification prototype built using C# and .NET at Microsoft's Annual TechXchange program at their Hyderabad Campus on 11th and 12th of December 2008.
  • Microsoft Research India, Summer 2008
    Selected to work as a Research Intern at Microsoft Research India (Bangalore) under Dr. Amitav Das. The selection was based on my sixth semester Digital Signal Processing (DSP) project executed under his guidance. As part of the internship, worked on various signal processing and classification methods which were applied to problems in knowledge-based speech recognition, character recognition and user-identification using face image and spoken password. Proprietary algorithms were developed in Matlab and some real-time implementations were made on C# and .NET platform. Two research papers were accepted in IEEE conferences (one national and one international) during this time.
  • Indian Institute of Science, Summer 2007
    Selected for the Summer Research Fellowship programme jointly sponsored by Indian Academy of Sciences (Bangalore), Indian National Science Academy (New Delhi) and National Academy of Sciences, India (Allahabad) in May-July 2007 and opted to work under Prof. L M Patnaik (Microprocessor Applications Laboratory, Super Computer Education and Research Centre, Indian Institute of Science, Bangalore). Systems for the automatic detection of epileptic activity in Electroencephalogram (EEG) signals were developed using Support Vector Machines and Artificial Neural Networks. Discrete wavelet transform was used for feature extraction from the EEG signals and a genetic algorithm was used to train the neural network. Implementations were done in Matlab on both Linux (RHEL 4) and Windows (Windows 2000) platforms. A 14 node high performance computational cluster was used to speed up processing. Necessary routines for distributed computing were developed and implemented through Matlab. A significant improvement in detection accuracy was achieved when compared to other automation attempts. The work has been reported in Article [1] above.

Some interesting stuff!

  • Article [1] above is currently ranked second among the Top 25 hottest articles in CMPB Journal (July to Sept. 2008)
  • I've got a DBLP page!! Not much on it yet, but hey, I just started!