Don't miss our weekly PhD newsletter | Sign up now Don't miss our weekly PhD newsletter | Sign up now

  Leveraging Vision Foundation Models for Zero-Shot Open Vocabulary 3D Scene Understanding

   Department of Computer Science

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
  Dr Muhammad Shahzad, Prof J Ferryman  No more applications being accepted  Competition Funded PhD Project (Students Worldwide)

About the Project

3D scene understanding aims to extract and interpret the semantic information of objects within their contextual environment. It is highly important and has wide range of diverse applications in different fields including robotics (e.g., autonomous navigation, terrestrial mapping, housekeeping, old-age assistance, agriculture), augmented/virtual reality, remote sensing (e.g., urban modeling, vegetation monitoring, surveying), 3D medical imaging, and many others. Existing methods leveraging recently proposed large vision language models have shown remarkable performance particularly in the 2D domain. Although few studies exist, but employing VFMs for 3D perceptions tasks remain largely unexplored. To bridge this gap, there is considerable promise in investigating the adaptation or extension of existing 2D vision foundation models for 3D perception tasks. Within this context, the research envisioned in this proposal aims to explore the zero-shot 3D scene understanding in open vocabulary setting. For this purpose, it is aimed to explore and adapt existing image-3D point cloud datasets. Using these datasets, the model development is intended to essentially focus on adapting existing vision models to process 3D data effectively and perform scene understanding tasks without relying on labeled data.

First degree in Computer Science with 2:1 or above MSc degree in the relevant subject areas is desired.

Enquiries for further details contact: Dr Muhammad Shahzad ([Email Address Removed])

Successful candidates will be paid an annual salary and stipend (£14,428 per annum) over the 4 year period and will have PhD fees covered at the Home level (Please note that students liable for international fees will need to pay the difference between these and the home fee rate). Fees for 2024/25 (amount payable each year) can be found at:

Link to advert:

Computer Science (8)

Funding Notes

The successful candidate will be employed as Graduate Teaching Assistant (GTA) at 0.25 FTE and as PhD student at 0.75 FTE for four years. Also, you do not need any research proposal to apply for this position. When applying, if the application system prompts you to submit a research proposal, please paste in ‘Computer Science GTA’ and move on to the next step in the application.

Where will I study?

Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.