CV0208: Internship - Anomaly Localization for Industrial Inspection

Mitsubishi•Cambridge, MA

58d•$6 - $8

About The Position

MERL is looking for a self-motivated intern to work on anomaly localization in industrial inspection setting using computer vision. The relevant topics in the scope include (but not limited to): cross-view image anomaly localization, how to train one model for multiple views and defect types, how to incorporate large foundation models in image anomaly localization, etc. The candidates with experiences of image anomaly localization in industrial inspection settings (e.g., MVTec-AD or VisA datasets) and usage of large foundation models are strongly preferred. The ideal candidate would be a PhD student with a strong background in computer vision and machine learning, and the candidate is expected to have published at least one paper in a top-tier computer vision, machine learning, or artificial intelligence venues, such as CVPR, ECCV, ICCV, ICML, ICLR, NeurIPS, or AAAI. Proficiency in Python programming and familiarity in at least one deep learning framework are necessary. The ideal candidate is required to collaborate with MERL researchers to develop algorithms and prepare manuscripts for scientific publications. The duration of the internship is ideally to be at least 3 months with a flexible start date.

Requirements

Proficiency in Python programming
familiarity in at least one deep learning framework
Experience with Python, PyTorch, and large foundation models (e.g. CLIP, ALIGN, etc.)

Nice To Haves

experiences of image anomaly localization in industrial inspection settings (e.g., MVTec-AD or VisA datasets) and usage of large foundation models
PhD student with a strong background in computer vision and machine learning
published at least one paper in a top-tier computer vision, machine learning, or artificial intelligence venues, such as CVPR, ECCV, ICCV, ICML, ICLR, NeurIPS, or AAAI