About

Pooja Venkatesan is a highly skilled video codec engineer with over 2.5 years of industry experience in the video compression domain. She has a deep understanding of the HEVC standard, FFmpeg, and GStreamer, and has hands-on experience in the development of some of the niche algorithms in open-source HEVC encoder x265. Currently working as a Software Engineer at Multicoreware, Pooja has contributed significantly to the development of x265, including the design and development of an in-built scaler to scale input to multiple different resolutions. She has also worked on integrating client's HW-based HEVC, H264, VP9, and JPEG decoders to FFmpeg and Gstreamer application and has been involved in the end-to-end design and development of a new framework that performs batch decoding of frames and is further fed to the classification/inference engine. Pooja's past experience includes working as a Software Engineer Intern at Multicoreware, where she worked on improving the temporal metric in video quality metric VMAF. She devised an algorithm to compute motion in a video involving block classification, clustering, and computing motion scores. This temporal feature is fed along with other spatial features of VMAF into a Support Vector Regression framework where the model is trained. On average, it achieves a better correlation to opinion score when compared to the existing VMAF by 8.7%. Pooja holds a Bachelor's degree in Computer Science from RMK Engineering College and has completed her 12th from DAV School, Gopalapuram, and her till 10th from Kendriya Vidyalaya. She is a highly skilled software engineer with 5.19 years of relevant experience and is always interested in experimenting and learning more.