Better Compression with Deep Pre-Editing
arXiv:2002.00113 · doi:10.1109/TIP.2021.3096085
Abstract
Could we compress images via standard codecs while avoiding visible artifacts? The answer is obvious -- this is doable as long as the bit budget is generous enough. What if the allocated bit-rate for compression is insufficient? Then unfortunately, artifacts are a fact of life. Many attempts were made over the years to fight this phenomenon, with various degrees of success. In this work we aim to break the unholy connection between bit-rate and image quality, and propose a way to circumvent compression artifacts by pre-editing the incoming image and modifying its content to fit the given bits. We design this editing operation as a learned convolutional neural network, and formulate an optimization problem for its training. Our loss takes into account a proximity between the original image and the edited one, a bit-budget penalty over the proposed image, and a no-reference image quality measure for forcing the outcome to be visually pleasing. The proposed approach is demonstrated on the popular JPEG compression, showing savings in bits and/or improvements in visual quality, obtained with intricate editing effects.
References in corpus (11)
- Adam: A Method for Stochastic Optimization
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Instance Normalization: The Missing Ingredient for Fast Stylization
- Variational image compression with a scale hyperprior
- End-to-end Optimized Image Compression
- NIMA: Neural Image Assessment
- CAS-CNN: A Deep Convolutional Neural Network for Image Compression Artifact Suppression
- Rethinking Lossy Compression: The Rate-Distortion-Perception Tradeoff
- Postprocessing of Compressed Images via Sequential Denoising
- Optimized Pre-Compensating Compression