Skip to content
Hallway
0:00 / 0:00

#cvpr#rawimages#low-lightenhancement

CVPR 2018: Learning to See in the Dark

Chen Chen

239 subscribers • 1 video on Hallway

YouTube
  • Recommended by 1 person: 1 on Hacker News.
  • Linked in the descriptionarXiv 1805.01934
  • May be out of date. Uploaded 8 years ago, in an area where tools change fast.

Intermediate. Assumes working knowledge.

190K views8 years ago1 recommendation

AI and machine learningComputer visionNeural networks and deep learning

This video demonstrates a neural network trained on raw sensor data to produce sharp, well-exposed images from extremely low-light environments.

Paper: arxiv.org/abs/1805.01934 Project page: cchen156.web.engr.illinois.edu/SID.html

Why it's here

1 person recommends this

What people say. Recommended as a remarkable demonstration of machine learning outperforming traditional digital signal processing for low-light image enhancement.

ispon Hacker News • Recommends

On “Learning to See in the Dark”

Remarkable machine learning result for "producing astoundingly sharp photos in very low light" (Cory Doctorow - link ) Demo example (one of many examples - drag the middle slider from left-to-right): link GitHub: link Paper: link Video: link to this video
View on Hacker News

lovelearningon Hacker News • Mentions

On “Why OpenCV uses BGR color space (2015)”

…ge enhancement1 - they found that training on raw images gives much better enhancement than JPEG. The data availability problem may start getting solved in the near future because mobiles (atleast Android) support storing and processing raw images, and on-device learning and inference capabilities are getting better all the time on mobiles. 1: link to this video…
View on Hacker News

peraon Hacker News • Mentions

On “Learning to See in the Dark (2018)”

To me the results seem vastly superior to those sort of simple DSP algorithms. The video shows a comparison with some denoising: link to this video
View on Hacker News