Sriram Nagaswamy

Build and manage research/commercialization teams based on project requirements. Define product roadmap using research reports, market trends and data analysis. Interact and manage global teams. Major technologies: Machine learning/Data mining and computer vision on embedded platforms (Android) Paper published in Journal of Computer and communications and accepted in the International Data mining Conference '15 A Bayesian Approach to Identify Photos Likely to Be More Popular in Social Media A patent filed with USPTO.

Setup and maintain our cloud based multi-conference unit (MCU), for the desktop and box based VC service. Implemented error resilience mechanisms such as FEC and LPR. Designed and implemented video codecs for the main video conferencing system.

Involved in documentation of vision and scope as well as functional and design documents. Proficient in working with additional TI hardware peripherals like iMX, VLD, LPF, Sequencer etc. Experienced in implementing assembly coding for C5x and C6x TI DSP’s. Experienced in designing regression testing environment for automated testing. Major Projects H.264 baseline profile 720p decoder and HP/MP D1 decoder on OMAP3 clocked at 430MHz. Currently used in Motorola's leading smart phones. H.264 baseline profile 720p decoder and D1 MP/HP decoder on DaVinci (DM420). The HP decoder is used in Verismo’s VuNow TV box Implemented a constant bit rate rate control algorithm on low the latency Simple Profile encoder on the DM320 platform. Used in surveillance systems. Designed and implemented the Motion compensation, Intra Prediction, DCT, Quantization, inverse quantization, IDCT, reconstruction, future SAD and variance for the H.264 720p encoder on Davinci.