Avinab Saha
Research Scientist at Google
Mountain View, California, United States
Summary
👤
Senior🎓
Top SchoolAvinab Saha is a research scientist at Google Research with a decade of experience bridging machine learning, visual perception, and applied image/video systems. He completed a PhD at UT Austin where he advanced self/semi-supervised representation learning and perceptual quality metrics for video, VR/AR, and mobile cloud gaming, and interned on diffusion models and RLHF at Google. Prior to grad school he built commercial AI features for Samsung flagship phones and 8K TVs, shipping video compression and gallery zoom enhancements to millions of users. Comfortable moving ideas from papers to product, he pairs deep technical rigor (4.0 PhD, top IIT Kharagpur honors) with hands-on deployment experience on constrained devices. He’s particularly focused on self-supervised and generative methods for visual perception, with a knack for translating neuroscience-inspired insights into practical video and display technologies.
10 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy - PhD, Electrical and Computer Engineering, 4.0, Doctor of Philosophy - PhD, Electrical and Computer Engineering, 4.0 at The University of Texas at AustinBachelor of Technology (B.Tech.), Major in Electronics and Electrical Communication Engineering, Minor in Computer Science and Engg, 9.12/10, Bachelor of Technology (B.Tech.), Major in Electronics and Electrical Communication Engineering, Minor in Computer Science and Engg, 9.12/10 at Indian Institute of Technology, Kharagpur
English, Hindi, Bengali
Stackoverflow
1reputation
65reached
0answers
1question
Github Skills (71)
perceptual10
computer-vision10
fusion10
image-quality10
video-quality10
ai9
python9
opencv9
deep-learning9
residual-networks9
feature-detection9
machine-learning9
contrastive-learning9
unsupervised-learning9
video-processing8
Programming languages (5)
Github contributions (5)
avinabsaha/ReIQA
Aug 2023 - Aug 2024
Official implementation for CVPR2023 Paper "Re-IQA : Unsupervised Learning for Image Quality Assessment in the Wild"
Contributions:4 PRs, 4 pushes, 14 comments in 1 year
blind-image-quality-assessmentcontrastive-learningcvprcvpr2023fr-iqa
avinabsaha/HIDRO-VQA
Dec 2023 - Jul 2024
Official Implementation of WACV 2024 Paper "HIDRO-VQA : High Dynamic Range Oracle for Video Quality Assessment"
Contributions:7 PRs, 10 pushes, 1 comment in 7 months
contrastive-learninghdrself-supervised-learningvideovideo-quality-assessment
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.