Hervé Jégou
Also known as Herve Jegou
Jégou earned a Rennes PhD in 2005 on resilient compression and source-channel coding. He joined Inria Grenoble’s LEAR group in 2006 and moved to Inria Rennes in 2009. His research applies product quantization to search in compressed vector collections. It focuses on indexing images, video and vectors at scales from millions to billions without first decompressing the entire collection. He moved to FAIR as a research scientist in May 2015.
- Now at
- FAIR (Meta)
- Works
- 272
- Citations
- 36,967
- h-index
- 63
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2005–2009
From compression to computer vision
Jégou earned a Rennes PhD in 2005 on resilient compression and source-channel coding. He joined Inria Grenoble’s LEAR group in 2006 and moved to Inria Rennes in 2009.
-
As of October 2026
Searching compressed vectors
Jégou’s research applies product quantization to search in compressed vector collections. It focuses on indexing images, video and vectors at scales from millions to billions without first decompressing the entire collection.
-
2015
Move to FAIR
He moved to FAIR as a research scientist in May 2015.
Sources and credits
Photo: people.rennes.inria.fr; creator not stated. License: Not stated by source. Source: people.rennes.inria.fr. Cropped to a square and resized. Ask us to remove or replace this photo.