Research Fellow/Engineer (AI On Video Analytics) - DW(1074) - Part-Time

About the Project

The SITx NVIDIA AI Centre (SNAIC) at the Singapore Institute of Technology(SIT) is seeking a highly motivated Research Fellow/Engineer to join a strategic collaboration with Procter & Gamble (P&G).

This project aims to develop next-generation AI technologies for concept-driven video understanding from consumer-recorded facial care videos. Unlike conventional action recognition, the research focuses on understanding high-level semantic concepts, behavioural patterns, product usage, and skin care routines through advanced Vision AI, Multimodal AI, and Foundation Models.

The successful candidate will work closely with researchers from SNAIC, NVIDIA, and P&G to develop novel AI methodologies with direct industrial impact.

Responsibilities

The successful candidate will:

• Conduct cutting-edge research in Computer Vision, Video Understanding, and Multimodal AI.
• Design AI models for concept-driven video understanding of consumer facial care behaviours.
• Work with PI and company to develop the AI solution
• Develop novel algorithms for:

-     Fine-grained video understanding

-     Concept learning

-     Temporal reasoning

-     Vision-Language Models (VLMs)

-     Multimodal representation learning

-     Self-supervised and weakly supervised learning

• Fine-tune and adapt Foundation Models and Vision-Language Models for domain-specific applications.
• Build end-to-end AI pipelines for video analytics and deployment.
• Work with industrial datasets and collaborate with domain experts from P&G.
• Publish research findings in top-tier AI conferences and journals.
• Mentor graduate students and support research activities within SNAIC.

Required Qualifications

Applicants should possess:

• PhD/Ms/BSc in Computer Science, Artificial Intelligence, Electrical Engineering, or a related discipline.
• Strong research background in one or more of:

-     Computer Vision

-     Machine Learning

-     Deep Learning

-     Video Understanding

-     Multimodal AI

• Excellent programming skills in Python.
• Hands-on experience with:

-     PyTorch

-     Hugging Face Transformers

-     OpenCV

• Experience working with deep learning models for image or video analysis.