ABCD: Arbitrary Bitwise Coefficient for De-Quantization

Woo Kyoung Han, Byeonghun Lee, Sang Hyun Park, Kyong Hwan Jin; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023, pp. 5876-5885

Abstract

Modern displays and contents support more than 8bits image and video. However, bit-starving situations such as compression codecs make low bit-depth (LBD) images (<8bits), occurring banding and blurry artifacts. Previous bit depth expansion (BDE) methods still produce unsatisfactory high bit-depth (HBD) images. To this end, we propose an implicit neural function with a bit query to recover de-quantized images from arbitrarily quantized inputs. We develop a phasor estimator to exploit the information of the nearest pixels. Our method shows superior performance against prior BDE methods on natural and animation images. We also demonstrate our model on YouTube UGC datasets for de-banding. Our source code is available at https://github.com/WooKyoungHan/ABCD

Related Material

[pdf] [supp]
[bibtex]
@InProceedings{Han_2023_CVPR, author = {Han, Woo Kyoung and Lee, Byeonghun and Park, Sang Hyun and Jin, Kyong Hwan}, title = {ABCD: Arbitrary Bitwise Coefficient for De-Quantization}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2023}, pages = {5876-5885} }