Vijay Badrinarayanan

Legacy & significance

Badrinarayanan helped write one of computer vision's most-cited deep learning papers, SegNet, before moving into frontier AI research at DeepMind, and now leads AI development at Wayve, applying end-to-end learned driving models rather than the hand-coded rule systems most autonomous vehicle companies rely on.

Major achievements & key works

• Co-authoring SegNet (2015). Introduced an encoder-decoder deep learning architecture for pixel-wise image segmentation that became a standard reference architecture across computer vision research and industry applications.
• Research scientist, DeepMind. Contributed to frontier deep learning research at one of the field's leading labs.
• VP of AI, Wayve. Leads development of Wayve's end-to-end learned driving models, an approach that trains a single AI system to drive directly from sensor input rather than composing separate hand-engineered modules.
• PhD research in Bayesian computer vision, University of Cambridge. Built the academic foundation underlying his later deep learning segmentation and perception work.

Pivotal turning points

Year	Event	Impact
2015	Publishes SegNet with collaborators at Cambridge	Introduces a widely adopted segmentation architecture
2016	Joins DeepMind as a research scientist	Moves into frontier AI research
2021	Joins Wayve as VP of AI	Leads end-to-end AI development for autonomous driving

Sources & further reading

Primary sources

• SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation, Badrinarayanan, Kendall, Cipolla, IEEE TPAMI (2017)
• Vijay Badrinarayanan, Wayve leadership team, official site

Recommended secondary sources