In this section

Cognitive Vision in Robotic Surgery

The Cognitive Vision in Robotic Surgery Lab is developing computer vision and AI techniques for intraoperative navigation and real-time tissue characterisation.

Head of Group

Dr Stamatia (Matina) Giannarou

411 Bessemer Building
South Kensington Campus

+44 (0) 20 7594 8904

What we do

Surgery is undergoing rapid changes driven by recent technological advances and our on-going pursuit towards early intervention and personalised treatment. We are developing computer vision and Artificial Intelligence techniques for intraoperative navigation and real-time tissue characterisation during minimally invasive and robot-assisted operations to improve both the efficacy and safety of surgical procedures. Our work will revolutionize the treatment of cancers and pave the way for autonomous robot-assisted interventions.

Why it is important?

With recent advances in medical imaging, sensing, and robotics, surgical oncology is entering a new era of early intervention, personalised treatment, and faster patient recovery. The main goal is to completely remove cancerous tissue while minimising damage to surrounding areas. However, achieving this can be challenging, often leading to imprecise surgeries, high re-excision rates, and reduced quality of life due to unintended injuries. Therefore, technologies that enhance cancer detection and enable more precise surgeries may improve patient outcomes.

How can it benefit patients?

Our methods aim to ensure patients receive accurate and timely surgical treatment while reducing surgeons' mental workload, overcoming limitations, and minimizing errors. By improving tumor excision, our hybrid diagnostic and therapeutic tools will lower recurrence rates and enhance survival outcomes. More complete tumor removal will also reduce the need for repeat procedures, improving patient quality of life, life expectancy, and benefiting society and the economy.

Meet the team

Dr Stamatia Giannarou
Senior Lecturer

Dr Po Wen Lo
Research Associate

Mr Alfie Roddan
Research Postgraduate

Mr Alistair G Weld
Research Postgraduate

Mr Chi Xu
Research Assistant

Mr Haozheng Xu
Research Assistant in Surgical Robot Vision

Mr Yihang Zhou
Research Assistant

Citation

BibTex format

@inproceedings{Huang:2022:10.1109/CVPRW56347.2022.00492,
author = {Huang, B and Zheng, J-Q and Giannarou, S and Elson, DS},
doi = {10.1109/CVPRW56347.2022.00492},
pages = {4459--4466},
publisher = {IEEE},
title = {H-Net: unsupervised attention-based stereo depth estimation leveraging epipolar geometry},
url = {http://dx.doi.org/10.1109/CVPRW56347.2022.00492},
year = {2022}
}

Download

RIS format (EndNote, RefMan)

TY  - CPAPER
AB  - Depth estimation from a stereo image pair has become one of the most explored applications in computer vision, with most previous methods relying on fully supervised learning settings. However, due to the difficulty in acquiring accurate and scalable ground truth data, the training of fully supervised methods is challenging. As an alternative, self-supervised methods are becoming more popular to mitigate this challenge. In this paper, we introduce the H-Net, a deep-learning framework for unsupervised stereo depth estimation that leverages epipolar geometry to refine stereo matching. For the first time, a Siamese autoencoder architecture is used for depth estimation which allows mutual information between rectified stereo images to be extracted. To enforce the epipolar constraint, the mutual epipolar attention mechanism has been designed which gives more emphasis to correspondences of features that lie on the same epipolar line while learning mutual information between the input stereo pair. Stereo correspondences are further enhanced by incorporating semantic information to the proposed attention mechanism. More specifically, the optimal transport algorithm is used to suppress attention and eliminate outliers in areas not visible in both cameras. Extensive experiments on KITTI2015 and Cityscapes show that the proposed modules are able to improve the performance of the unsupervised stereo depth estimation methods while closing the gap with the fully supervised approaches.
AU  - Huang,B
AU  - Zheng,J-Q
AU  - Giannarou,S
AU  - Elson,DS
DO  - 10.1109/CVPRW56347.2022.00492
EP  - 4466
PB  - IEEE
PY  - 2022///
SP  - 4459
TI  - H-Net: unsupervised attention-based stereo depth estimation leveraging epipolar geometry
UR  - http://dx.doi.org/10.1109/CVPRW56347.2022.00492
UR  - https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000861612704057&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=a2bf6146997ec60c407a63945d4e92bb
UR  - https://ieeexplore.ieee.org/document/9856940
UR  - http://hdl.handle.net/10044/1/102069
ER  -