Keywords: Computational imaging, hardware prototyping, 3D printing.
GitHub project page: https://github.com/LCAV/LenslessPiCam
Medium blog posts: https://go.epfl.ch/lenslesspicam
Contact: Eric Bezzam
Levels: Bachelor / Master
Sections: IC, EE, ME, MT
Description: Lensless imaging throws away centuries-old notions of taking pictures. By shifting the image formation from the lens to the digital post-processing, the constraints imposed by lenses (e.g. focusing distances, weight, form factor, cost, etc) can be largely relaxed. Consequently, a thin, light, and potentially inexpensive optical element can be used with an appropriately designed computational algorithm to perform imaging (see above).
The goal of this project is to fabricate such optical elements from cheap materials and with accessible techniques, namely epoxy and 3D printing, such that they can be used with a working prototype. Designing a mount to fix the mask in front of the camera sensor is also of interest.
Profile:
- Experience with designing STL files for 3D printing.
- Comfortable with Python.
- (Optional) experience with one of the workshops at EPFL for 3D printing and CNC milling (SPOT or AFA).
- (Optional) experience with epoxy.
Example masks:
- Pseudo-random binary amplitude mask: https://arxiv.org/abs/1509.00116
- Multifocal microlens array: https://arxiv.org/abs/2006.16343
- High-contract contour phase mask: https://ieeexplore.ieee.org/document/9076617
- Fresnel zone aperture: https://en.wikipedia.org/wiki/Zone_plate