PhD Scholarship (Industry) - Privacy-Preserving AI for Human-Centric Video Analytics

This opportunity was shared with us by Dr. Deval Mehta, the project supervisor at Monash University. Monash University, in collaboration with the Australian AI company Aervision Pty Ltd, is opening applications for a fully funded Industry Linked PhD Scholarship focused on building the next generation of privacy-preserving artificial intelligence for human-centric video analytics. The position is based at Monash's Clayton campus in the Department of Data Science & AI, Faculty of Information Technology, and is backed by the Australian Government National Industry PhD Program.

This is a rare chance for a computer vision or machine learning researcher to work at the intersection of academic research and commercial AI product development, with direct access to real-world industry datasets and Aervision's deployed video analytics platforms.

Overview of the Opportunity

Video analytics is increasingly used across critical infrastructure, public safety, and smart environments, but most existing systems face a core tension: they need to understand complex human behaviour accurately while also protecting the privacy of the people being observed, and they need to do this in real time. This PhD Scholarship project sets out to resolve that tension by developing foundation models that are accurate, adaptable to new environments without heavy retraining, and privacy-safe by design.

The successful candidate will be supervised by Dr Deval Mehta and will split their time between Monash's research environment and Aervision's research and development team, giving them exposure to how academic AI research gets translated into deployable commercial products.

Scholarship Funding and Financial Support

One of the strongest draws of this opportunity is its layered funding structure, which is designed to make full-time doctoral research financially sustainable while also rewarding the industry-linked nature of the project. The funding package includes:

1. A Research Living Allowance currently valued at $37,145 AUD per annum (2026 rate, with annual indexation) for the duration of the PhD
2. A Faculty of Information Technology Tuition Fee Scholarship, available to eligible international students
3. An Industry Top-up Scholarship worth $10,000 per annum, funded through the Aervision partnership
4. FIT Candidature Funding of $4,000 for the full duration of candidature
5. A one-off travel grant of up to $1,265 from the Monash Graduate Research Office
6. A Top-up Government Scholarship of $7,135 per annum

Combined, these components create a competitive, multi-source scholarship package that goes beyond a standard stipend, reflecting the project's dual academic-industry funding model.

Why This Project Matters: The Research Problem

Modern video analytics systems are expected to do three things simultaneously: interpret nuanced human behaviour, protect the identity and privacy of the individuals captured on camera, and operate reliably in real time on deployed hardware. Most current approaches fall short on at least one of these fronts. Systems that are highly accurate at behaviour recognition often rely on identifiable visual data, raising privacy and compliance concerns. Systems designed with privacy safeguards in mind, on the other hand, frequently lose accuracy or struggle to generalise to new environments without significant retraining.

This PhD project is designed to close that gap by developing scalable, adaptable AI models that maintain strong behavioural understanding while using anonymised human representations rather than raw identifiable footage.

Core Research Directions

Candidates will have the opportunity to explore multiple interconnected research threads, including:

1. Human action and activity recognition
2. Individual and group behaviour understanding
3. Long-term behaviour prediction
4. Multi-task learning and foundation model design
5. Zero-shot and open-set recognition
6. Privacy-preserving AI built on anonymised human representations
7. Efficient AI deployment on edge devices
8. Applied AI systems for security, healthcare, and smart environment use cases

The research will draw on anonymised industry datasets supplied through the Aervision partnership and will be validated directly on Aervision's commercial AI platforms. This gives the project a genuine pathway from theoretical research to real-world deployment, an unusually direct pipeline for a doctoral program. It also positions the work as a contribution to Australia's broader push for trustworthy, responsible, and commercially viable AI systems.

 

Who Should Apply: Candidate Profile

Monash and Aervision are looking for candidates with a solid grounding in computer science, data science, or artificial intelligence, along with genuine enthusiasm for both rigorous academic research and hands-on industry collaboration.

Essential Skills and Experience

Applicants should be able to demonstrate capability in:

1. Deep learning and machine learning fundamentals
2. Computer vision techniques
3. Python programming
4. Experience with PyTorch or TensorFlow
5. Strong mathematical and analytical reasoning

Desirable Skills and Experience

The following experience is not mandatory but will strengthen an application:

1. Video understanding and human action recognition
2. Vision-language models or other foundation model architectures
3. Prior work in privacy-preserving AI
4. Experience training large-scale models
5. Publication history in top-tier AI venues such as CVPR, ICCV, AAAI, or NeurIPS

Strong written and verbal communication skills are also valued highly, given the project's dual academic and commercial audience.

Academic Eligibility Requirements

To be considered for admission into the PhD program, applicants must meet one of the following academic qualification criteria:

1. A bachelor's degree of at least four years in a relevant discipline that includes a research thesis or project, with an overall average equivalent to First Class Honours; or
2. A master's degree in a relevant discipline that includes a research thesis or project component equal to at least 25 percent of one year of full-time study, with an overall average equivalent to First Class Honours; or
3. A qualification, or a combination of qualifications and relevant professional experience, assessed as equivalent by Monash's Graduate Research Committee (or its delegate).

Monash University has stated its ongoing commitment to diversity, equity, fairness, and openness, and supports the gender equity principles set out in the Athena SWAN Charter.

What the Scholarship Offers Beyond Funding

Beyond the direct financial support, the scholarship provides several career-building advantages that are worth weighing carefully:

1. A prestigious Australian Government Industry PhD Scholarship
2. Direct, sustained collaboration with an innovative Australian AI company
3. Access to large-scale, real-world industry datasets not typically available to academic researchers
4. An embedded industry placement running throughout the candidature
5. Use of Monash University's high-performance computing infrastructure
6. Strong opportunities to publish in leading AI venues, including CVPR, ICCV, AAAI, and NeurIPS
7. Formal training in research commercialisation and technology translation
8. Solid employment prospects across both academic and industry AI careers on graduation

For a researcher weighing where to invest 3.5 years of doctoral study, the combination of applied industry exposure and academic publication opportunity is a distinctive feature of this program compared with a standard, university-only PhD track.

Application Process: A Two-Stage Selection

Unlike a standard PhD application, this position runs through a two-stage selection process designed to assess both the candidate's research thinking and their fit with the project before a full application is submitted.

Stage 1: Expression of Interest (EOI)

Interested applicants should send an Expression of Interest directly to Dr Deval Mehta at deval.mehta@monash.edu. The EOI should include:

1. A current CV
2. Academic transcripts
3. A cover letter
4. A draft research proposal of up to 5 pages, outlining a proposed approach to building privacy-preserving AI for human-centric video analytics

The draft research proposal is the most substantial component of Stage 1. It should clearly articulate the current limitations in video analysis research, propose a framework or approach to address those limitations, and highlight genuine elements of novelty and innovation. It should also explain the applicant's motivation for pursuing a PhD within the Department of Data Science & AI and why this specific project appeals to them.

Stage 2: Discussion and Full Application

Candidates who progress past the EOI stage will be invited to discuss their research ideas directly with Dr Deval Mehta. Following this conversation, successful candidates will be supported in developing and submitting a full PhD application.

Applications close on Sunday, 30 August 2026, at 11:55pm AEST. Importantly, Monash has indicated that shortlisting and interviews will begin as soon as suitable applications are received, rather than waiting for the closing date. Early applicants are therefore strongly encouraged to submit their EOI as soon as they are ready, rather than waiting until the deadline approaches.

Key Contact Information

1. Enquiries: Dr Deval Mehta - deval.mehta@monash.edu
2. Job Reference Number: 696616
3. Campus: Clayton, Monash University
4. Industry Partner: Aervision Pty Ltd

Explore More Scholarship Opportunities

If this Monash PhD Scholarship doesn't quite match your field or study plans, it's worth browsing a few other funded opportunities we've covered that might be a better fit. For those specifically drawn to research careers in Europe, the Munich International Youth Fellowships 2027 in Germany is another strong option worth reviewing.

Students who are still weighing IELTS-free pathways may also want to look at our guides on how to study in Poland without IELTS on a scholarship, study in Germany without IELTS on a scholarship, and study in Singapore without IELTS on a scholarship, all of which open doors for applicants who don't have an English proficiency test on hand yet.

If your interests lean toward the Middle East or Northern Europe instead, take a look at our roundup of the best Saudi Arabia scholarships to study in Saudi Arabia, or our detailed breakdown of the cost of studying in Finland along with available scholarships, which is especially useful if affordability is a major factor in your decision.

Final Thoughts

This scholarship represents a genuinely rare structure in doctoral research: a fully funded, government-backed Industry PhD that pairs deep academic supervision with hands-on placement inside a working AI company. For candidates interested in computer vision, foundation models, and the growing field of responsible, privacy-preserving AI, this project offers both intellectual challenge and a concrete route from research to real-world deployment. Given the rolling shortlisting process, candidates who are eligible and interested should treat the application timeline as a priority and prepare their research proposal early.

Official Website

Link: https://careers.pageuppeople.com/513/cw/en/job/696616/phd-scholarship-industry-privacypreserving-ai-for-humancentric-video-analytics