Enhance dark images in new untrained domain by simulating image pipeline for training
Enhance dark images in new untrained domain by simulating image pipeline for training
Towards Robust Low Light Image Enhancement
arXiv paper abstract https://arxiv.org/abs/2205.08615v1
arXiv PDF paper https://arxiv.org/pdf/2205.08615v1.pdf
… study the problem of making brighter images from dark images found in the wild.
… suffer from color shifts caused by quantization and from sensor noise.
… don’t know the true camera reponse function for such images and they are not RAW.
… use a supervised learning method, relying on a straightforward simulation of an imaging pipeline to generate usable dataset for training and testing.
… approach outperforms the state of the art quantitatively.
Qualitative comparisons suggest strong improvements in reconstruction accuracy.
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
