Ronnie Clark

Ronnie Clark is an Associate Professor of Computer Science at Oxford whose research spans 3D computer vision, graphics and neural rendering. He earned an MSc in Information Engineering from the University of the Witwatersrand and a DPhil at Oxford, joined Imperial's Dyson Robotics Lab in 2017, and received a CVPR Best Paper Honourable Mention in 2018. His early robotics work included SemanticFusion, a 2017 dense RGB-D mapping system, and his 2022 paper addressed online photorealistic scene capture. In 2025, he collaborated with Yuanze Lin and Microsoft on Olympus, a computer vision framework that uses a multimodal language model to route tasks.

Now at: University of Oxford

Timeline

2017

SemanticFusion at TUM

The TUM Smart Robotics Lab lists him as a former collaborator on SemanticFusion, which adds a semantic layer to a real-time dense RGB-D SLAM system. The lab's related paper on dense 3D semantic mapping appears at ICRA in 2017.
2017–2020

Dyson Robotics Lab

He earns an MSc in Information Engineering from the University of the Witwatersrand and a DPhil from Oxford. From 2017 to 2020, he is a member of Imperial College London's Dyson Robotics Lab; its news lists him among recent fellows.
2018

CVPR Best Paper mention

He receives a Best Paper Honourable Mention at CVPR in 2018.
2019

Deep learning and geometry

At the BMVA's Geometry and Deep Learning meeting in 2019, he presents “Deep Learning and Geometry: A Top-Down Approach.”
2022

Associate Professor at Oxford

After an Imperial College Research Fellowship, he joins the University of Oxford in 2022 as an Associate Professor of Computer Science.
2022

Photorealistic scene capture

He publishes Volumetric Bundle Adjustment for Online Photorealistic Scene Capture at CVPR in 2022.
2025

Olympus routes vision tasks

He and DPhil student Yuanze Lin collaborate with Microsoft to introduce Olympus. The framework uses a multimodal large language model to interpret user instructions, route tasks to specialized models, and handle up to 20 simultaneous tasks with action chains of up to five steps.
As of October 2026

3D vision at Oxford

As of October 2026, he is an Associate Professor of Computer Science at Oxford. His research spans 3D computer vision and graphics, including neural rendering, dense visual SLAM and scene understanding.