A U-net reconstructs MNIST, EMNIST, and Kanji49 images from raw sensor data of a see-through lensless camera, but classification benefits are inconsistent and the manuscript is incomplete.
Computational snapshot angular-spectral lensless imaging
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
By placing a diffractive element in front of an image sensor, we are able to multiplex the spectral and angular information of a scene onto the image sensor. Reconstruction of the angular-spectral distribution is attained by first calibrating the angular-spectral response of the system and then, applying optimization-based matrix inversion. In our proof-of-concept demonstration, we imaged the 1D angle and the spectrum with resolutions of 0.15o and 6nm, respectively. The information is reconstructed from a single frame, thereby enabling snapshot functionality for video-rate imaging.
fields
eess.IV 1years
2019 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Machine-learning enables Image Reconstruction and Classification in a "see-through" camera
A U-net reconstructs MNIST, EMNIST, and Kanji49 images from raw sensor data of a see-through lensless camera, but classification benefits are inconsistent and the manuscript is incomplete.