Fabien Racapé

Palo Alto, California, United States

Join Prog.AI to see contacts

Summary

🤩

Rockstar

🎓

Top School

Fabien Racapé is a researcher at Technicolor AI Labs with a PhD in signal and image processing and over a decade of experience in video compression, now pivoting to deep-learning-driven methods for image, 3D and 360° video coding. He combines academic rigor—from ENSPG, Grenoble INP and INSA Rennes—with hands-on research stints at Fraunhofer and INRIA, and a long tenure at Technicolor focusing on perceptual quality and standards. Based in Palo Alto, he blends algorithm development with practical tooling, having improved build and benchmarking workflows for the popular CompressAI project to add HM and AV1 codec support and more robust error reporting. Known for bridging theory and implementation, he brings a rare mix of physics-trained analytical thinking and pragmatic engineering to complex multimedia challenges.

5 years of coding experience

Doctor of Philosophy (PhD), Signal and image processing, Doctor of Philosophy (PhD), Signal and image processing at INSA Rennes

Master of Science (MS), Signal processing, image processing, speech processing, Telecommunication, Master of Science (MS), Signal processing, image processing, speech processing, Telecommunication at Grenoble Institute of Technology

Engineer degree, Physics; instruments, Engineer degree, Physics; instruments at ENSPG, Ecole Nationale Supérieure de Physique, Grenoble

German, English

Github Skills (10)

bash10

build-automation10

compression9

testing8

deep-neural-networks7

python7

pytorch7

machine-learning7

deep-learning7

neural-network7

Programming languages (1)

Python

Github contributions (5)

InterDigitalInc/CompressAI

Jun 2020 - Jan 2023

A PyTorch library and evaluation platform for end-to-end compression research

Role in this project:

DevOps Engineer

Contributions:25 reviews, 213 commits, 33 PRs in 2 years 8 months

Contributions summary:Fabien primarily focused on improving the build and testing processes within the project. Contributions included adding error reporting to the benchmark script to identify codec installation issues. They also modified paths to external dependencies like VTM and TFCI and incorporated version checks. Furthermore, the user added the HM and AV1 codecs to the benchmark script to add support for these codecs.

pytorchpythonend-to-enddeep-learningcompression

InterDigitalInc/CompressAI-Trainer

Mar 2023 - Sep 2024

Contributions:7 pushes in 1 year 5 months

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.

Request Free Trial