CV0050: Internship - Anomaly Localization for Industrial Inspection

Mitsubishi - Cambridge, MA

Full-time - Intern

Cambridge, MA

Transportation Equipment Manufacturing

About the position

MERL is seeking a self-motivated intern to focus on anomaly localization in industrial inspection using computer vision techniques. The role involves exploring various topics such as cross-view image anomaly localization, training models for multiple views and defect types, and integrating large foundation models into the anomaly localization process. The internship aims to contribute to the development of algorithms and scientific publications in the field of computer vision and machine learning.

Responsibilities

• Work on anomaly localization in industrial inspection settings using computer vision.
• Explore cross-view image anomaly localization techniques.
• Train models for multiple views and defect types.
• Incorporate large foundation models in image anomaly localization.
• Collaborate with MERL researchers to develop algorithms.
• Prepare manuscripts for scientific publications.

Requirements

• Currently pursuing a PhD in a relevant field such as computer vision or machine learning.
• Strong background in computer vision and machine learning.
• Experience with image anomaly localization in industrial inspection settings (e.g., MVTec-AD or VisA datasets).
• Proficiency in Python programming.
• Familiarity with at least one deep learning framework.
• Published at least one paper in top-tier venues (CVPR, ECCV, ICCV, ICML, ICLR, NeurIPS, or AAAI).

Nice-to-haves

• Experience with large foundation models in image anomaly localization.

Benefits

• Opportunity to work with leading researchers in the field.
• Gain hands-on experience in cutting-edge research.
• Flexible start date for the internship.