arXiv Open Access 2019

Extreme Image Coding via Multiscale Autoencoders With Generative Adversarial Optimization

Chao Huang Haojie Liu Tong Chen Qiu Shen Zhan Ma

Lihat Sumber

Abstrak

We propose a MultiScale AutoEncoder(MSAE) based extreme image compression framework to offer visually pleasing reconstruction at a very low bitrate. Our method leverages the "priors" at different resolution scale to improve the compression efficiency, and also employs the generative adversarial network(GAN) with multiscale discriminators to perform the end-to-end trainable rate-distortion optimization. We compare the perceptual quality of our reconstructions with traditional compression algorithms using High-Efficiency Video Coding(HEVC) based Intra Profile and JPEG2000 on the public Cityscapes and ADE20K datasets, demonstrating the significant subjective quality improvement.

Topik & Kata Kunci

eess.IV cs.LG

Penulis (5)

Chao Huang

Haojie Liu

Tong Chen

Qiu Shen

Zhan Ma

Format Sitasi

APA MLA BibTeX

Huang, C., Liu, H., Chen, T., Shen, Q., Ma, Z. (2019). Extreme Image Coding via Multiscale Autoencoders With Generative Adversarial Optimization. https://arxiv.org/abs/1904.03851

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2019
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓