Research Associate: 3D Object Detection for Robotics

Nanyang Technological University; Singapore; On-site; SGD 70,000 - 100,000; Full time

Job summary

NTU Singapore invites applications for a Research Associate to lead the computer vision work in a robotic pick-and-place project at the Hyundai-NTU-A*STAR Corp Lab. The role centers on developing 3D object detection and segmentation models and advancing data augmentation techniques.

The successful candidate will conduct experiments, analyze results, and contribute to publications while collaborating with engineers and researchers.

Qualifications

• Advanced knowledge of 2D/3D object detection and segmentation.
• Experience with SAM, foundation models, diffusion models, transfer learning and few-shot methods.
• Publications at top AI conferences (CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR) preferred.
• Strong research capabilities including hypothesis formulation and experimental design.
• Collaborative and proactive communication skills; ability to work independently or in a team.

Responsibilities

• Develop 3D objects and segmentation models for robotic pick and place project.
• Develop data augmentation/data synthesis methods to address limited training data.
• Develop algorithms to annotate 2D images through 3D data.
• Check data quality.
• Review weaknesses of existing methods for 3D object detection and segmentation.
• Carry out experiments and analyze results.
• Develop robotic pick and place with other researchers and engineers.
• Prepare reports and scientific papers.

Skills

3D object detection

3D segmentation

2D object detection

Foundation models

Diffusion models

Transfer learning

Few-shot learning

Deep learning

SAM

Robotics knowledge

Data augmentation

Education

Master's degree in Computer Engineering / Computer Science / Electronics Engineering

Job description

NTU Singapore invites applications for a Research Associate to lead the computer vision work in a robotic pick-and-place project at the Hyundai-NTU-A*STAR Corp Lab. The role centers on developing 3D object detection and segmentation models and advancing data augmentation techniques.

The successful candidate will conduct experiments, analyze results, and contribute to publications while collaborating with engineers and researchers.