cchen156/Learning-to-See-in-the-Dark
A TensorFlow implementation of a deep learning model that enhances extremely low-light images using raw sensor data from the See-in-the-Dark dataset.
Velocity · 7d
+1.9
★ / day
Trend
→steady
star history
This repository contains a TensorFlow implementation of the See-in-the-Dark paper from CVPR 2018. It trains a deep neural network to reconstruct well-exposed images from raw sensor data captured in low-light conditions. The model handles challenges of high ISO and short exposure photography in darkness by learning to reconstruct details from extremely underexposed inputs.