1 Jun 2026

Research Intern – Multimodal Computer Vision (RGB–Thermal Fusion)

Keenfinity Group

Eindhoven

1 Jun 2026

Apply

Research Intern position at Keenfinity focusing on multimodal computer vision, specifically RGB-thermal fusion for low-light environments. The intern will work on data handling, algorithm development, and evaluation frameworks in Eindhoven, Netherlands.

AIHybridInternshipInternshipPythonPyTorch

Research Intern – Multimodal Computer Vision (RGB–Thermal Fusion)

Apply

AIHybridInternshipInternshipPython

Salary

Not specified

Work Location

Eindhoven, North Brabant, Netherlands, NL

Work Model

Hybrid: option to work from home one day per week

Employment Type

Internship

Experience Level

Internship

Core Qualifications

Technical (Must-have)

PythonPyTorchTensorFlowKerascomputer visiondeep learning

Soft Skills

fluent in Englishteamworkinitiative

Preferred Qualifications

Technical (Nice-to-have)

multi-modal learningimage fusionlow-light visionimage registrationcalibrationoptical flowfeature matchingablation studies

Key Responsibilities

•Work with visible (RGB) and thermal image/video datasets
•Understand and reproduce baseline pipelines for image registration and fusion
•Design and implement a robust evaluation framework
•Implement and benchmark baseline RGB–thermal fusion methods
•Design and train a reliability-aware fusion model using attention or gated fusion
•Improve robustness for imperfect alignment and low-visibility scenarios
•Develop methods to identify and minimize artifacts like ghosting and distortions
•Evaluate approach against baselines using performance metrics and qualitative insights
•Validate improvements on downstream tasks such as object detection in low-light conditions (optional)
•Document work in a thesis/report and prepare a publication-ready results package

internshipcomputer visionmultimodalRGB-thermal fusiondeep learningPythonPyTorchEindhovenAIresearch

Key Responsibilities

•Work with visible (RGB) and thermal image/video datasets

•Understand and reproduce baseline pipelines for image registration and fusion

•Design and implement a robust evaluation framework

•Implement and benchmark baseline RGB–thermal fusion methods

•Design and train a reliability-aware fusion model using attention or gated fusion

•Improve robustness for imperfect alignment and low-visibility scenarios

•Develop methods to identify and minimize artifacts like ghosting and distortions

•Evaluate approach against baselines using performance metrics and qualitative insights

•Validate improvements on downstream tasks such as object detection in low-light conditions (optional)

•Document work in a thesis/report and prepare a publication-ready results package