Learning Content-Weighted Deep Image Compression

Li, Mu; Zuo, Wangmeng; Gu, Shuhang; You, Jane; Zhang, David

doi:10.1109/tpami.2020.2983926

Cited by 54 publications

(20 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Tecnick dataset is comprised of 100 uncompressed images with a resolution of 1200 × 1200. The learning-based methods are recent state-of-the-art methods, including Balle2017 [8], Li2018 [27], Lee2019 [32], Lee2020 [26], Li2020 [33], and Cheng2020 [34], We also compare with traditional methods, including VVC-Intra (4:4:4) [35], VVC-Intra (4:2:0), BPG (4:4:4) [3], JPEG2000, WebP [36], and JPEG. Both PSNR and MS-SSIM metrics are used.…”

Section: Methodsmentioning

confidence: 99%

Learned Image Compression with Discretized Gaussian-Laplacian-Logistic Mixture Model and Concatenated Residual Modules

Fu,

Liang,

Lin

et al. 2021

Preprint

View full text Add to dashboard Cite

Recently deep learning-based image compression methods have achieved significant achievements and gradually outperformed traditional approaches including the latest standard Versatile Video Coding (VVC) in both PSNR and MS-SSIM metrics. Two key components of learned image compression frameworks are the entropy model of the latent representations and the encoding/decoding network architectures. Various models have been proposed, such as autoregressive, softmax, logistic mixture, Gaussian mixture, and Laplacian. Existing schemes only use one of these models. However, due to the vast diversity of images, it is not optimal to use one model for all images, even different regions of one image. In this paper, we propose a more flexible discretized Gaussian-Laplacian-Logistic mixture model (GLLMM) for the latent representations, which can adapt to different contents in different images and different regions of one image more accurately. Besides, in the encoding/decoding network design part, we propose a concatenated residual blocks (CRB), where multiple residual blocks are serially connected with additional shortcut connections. The CRB can improve the learning ability of the network, which can further improve the compression performance. Experimental results using the Kodak and Tecnick datasets show that the proposed scheme outperforms all the state-of-the-art learning-based methods and existing compression standards including VVC intra coding (4:4:4 and 4:2:0) in terms of the PSNR and MS-SSIM. The project page is at https://github.com/fengyurenpingsheng/ Learned-image-compression-with-GLLMM.

show abstract

Section: Methodsmentioning

confidence: 99%

Learned Image Compression with Discretized Gaussian-Laplacian-Logistic Mixture Model and Concatenated Residual Modules

Fu,

Liang,

Lin

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In a similar spirit, Li et al [41] described a spatially adaptive bit allocation scheme, where the rate was estimated as the total number of codes allocated to different regions. They [42] further designed better relaxation strategies for learning optimal bit allocation.…”

Section: Learned Planar Image Compressionmentioning

confidence: 99%

Pseudocylindrical Convolutions for Learned Omnidirectional Image Compression

Ma¹,

Li²,

Zhang³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Although equirectangular projection (ERP) is a convenient form to store omnidirectional images (also known as 360°i mages), it is neither equal-area nor conformal, thus not friendly to subsequent visual communication. In the context of image compression, ERP will over-sample and deform things and stuff near the poles, making it difficult for perceptually optimal bit allocation. In conventional 360°image compression, techniques such as region-wise packing and tiled representation are introduced to alleviate the over-sampling problem, achieving limited success. In this paper, we make one of the first attempts to learn deep neural networks for omnidirectional image compression. We first describe parametric pseudocylindrical representation as a generalization of common pseudocylindrical map projections. A computationally tractable greedy method is presented to determine the (sub)-optimal configuration of the pseudocylindrical representation in terms of a novel proxy objective for rate-distortion performance. We then propose pseudocylindrical convolutions for 360°image compression. Under reasonable constraints on the parametric representation, the pseudocylindrical convolution can be efficiently implemented by standard convolution with the so-called pseudocylindrical padding.To demonstrate the feasibility of our idea, we implement an end-to-end 360°image compression system, consisting of the learned pseudocylindrical representation, an analysis transform, a non-uniform quantizer, a synthesis transform, and an entropy model. Experimental results on 19, 790 omnidirectional images show that our method achieves consistently better rate-distortion performance than the competing methods. Moreover, the visual quality by our method is significantly improved for all images at all bitrates.

show abstract

“…Various methods have been proposed to improve the rate-distortion trade-off. For example, [5,24,30,33,34,39] incorporate entropy prediction of learned representations based on context during training, and [25,26,30] employ importance maps internally for dynamic bit allocation of latent representations. Some approaches introduce additional models for hyper-prior, which provide side information for a conditional entropy model [6,24,32].…”

Section: Deep Image Compressionmentioning

confidence: 99%

Variable-Rate Deep Image Compression through Spatially-Adaptive Feature Transform

Song¹,

Choi²,

Han³

2021

Preprint

View full text Add to dashboard Cite

We propose a versatile deep image compression network based on Spatial Feature Transform (SFT) [47], which takes a source image and a corresponding quality map as inputs and produce a compressed image with variable rates. Our model covers a wide range of compression rates using a single model, which is controlled by arbitrary pixel-wise quality maps. In addition, the proposed framework allows us to perform task-aware image compressions for various tasks, e.g., classification, by efficiently estimating optimized quality maps specific to target tasks for our encoding network. This is even possible with a pretrained network without learning separate models for individual tasks. Our algorithm achieves outstanding rate-distortion trade-off compared to the approaches based on multiple models that are optimized separately for several different target rates. At the same level of compression, the proposed approach successfully improves performance on image classification and text region quality preservation via task-aware quality map estimation without additional model training. The code is available at the project website 1 .

show abstract

Learning Content-Weighted Deep Image Compression

Cited by 54 publications

References 20 publications

Learned Image Compression with Discretized Gaussian-Laplacian-Logistic Mixture Model and Concatenated Residual Modules

Learned Image Compression with Discretized Gaussian-Laplacian-Logistic Mixture Model and Concatenated Residual Modules

Pseudocylindrical Convolutions for Learned Omnidirectional Image Compression

Variable-Rate Deep Image Compression through Spatially-Adaptive Feature Transform

Contact Info

Product

Resources

About