As an intern you work on advanced multimodal computer vision problems, focused on improving visibility and robustness in low-light environments by combining RGB and thermal imaging.
Tasks:
- Work with visible (RGB) and thermal image/video datasets from internal and public sources
- Understand and reproduce baseline pipelines for image registration and fusion
- Design and implement a robust evaluation framework with quantitative metrics and visual assessment
- Implement and benchmark RGB-thermal fusion methods
- Design and train a confidence-aware fusion model using techniques such as attention or gated fusion
- Improve robustness for imperfect alignment and low-visibility scenarios
- Develop methods to identify and minimize artifacts such as ghosting, hallucinations and distortions
- Evaluate approach against baselines with performance metrics and qualitative insights
- Validate improvements on downstream tasks such as low-light object detection (optional)
- Document work in thesis/report and prepare publication-ready results