L. Venkata Subramaniam's Research Page

• Multimodal Speech Recognition and Animation

  Currently I am working on twin problems of multimodal speech recognition and audio driven facial animation. The audio and video provide orthogonal information in many cases and their combination aids in speech recognition greatly [MMSP 99][ICPR 00]. I am also working on the tracking of lips on a talking face and the extraction of visual features, for speech recognition, from the lip region [ICPR 00]. I am also working on audio driven facial animation using morphing based approaches [ICME 01]. Along with the Speech Team I have developed a translingual visual speech synthesis engine based on this approach [ICME 00].

• Underwater Image Processing

  I worked as part of the team developing algorithms and software for the Department of Electronics sponsored work on ADvanced Object VIsualization Techniques (ADOVIT) for the underwater scenario from August 1993 to September 1996 [ADOVIT 94][ADOVIT 95]. My work involved forming visual images using sonar data. Sonar data is very sparse and noisy. A multiframe imaging Technique developed by us is used to reduce speckle and noise [UUST 95]. Using the sonar data as hard constraints it is merged with a shape from shading model obtained of the scene from visual images. Segmentation and surface understanding techniques are then used to form a dense image for a human observer [UUST 95].

• I developed a sea-bottom visualization software package (jointly with Vinita Rai and R. Bahl) for the National Inst. of Oceanography (NIO), Goa. Data collected by a survey ship is processed to give visual images and contour maps of the sea bottom. The code for this is written in C. The graphics portions use Xlib and STARBASE libraries.

• Coding for AWGN and Fading Channels [Ph D Thesis].

  How Good is a code designed for the AWGN Channel? Tried to answer this question by obtaining a lower bound on the largest achievable rate vs Euclidean distance of the code. This work also suggests constellations in 2,3 and 4-dimensions over which asymptotically good codes may be found [IEEE Trans. IT][ISIT98][NCC 99]. Over Fading channels worked on TCM schemes with asymmetric PSK signal sets specially modeled to maximize performance [ISIT 98][IEEE Trans. VT 00].

• Channel modeling and simulation [Ph D Thesis]

  Modeling of the AWGN channel, satellite mobile channel, Ricean and Rayleigh fading channels, and the underwater acoustic channel to simulate the performance of various codes over them [OCEANS 98][IEEE Trans. VT 00][NSOE 97].