Image Compression Framework Research Articles

OoDHDR-Codec: Out-of-Distribution Generalization for HDR Image Compression

Recently, deep learning has been proven to be a promising approach in standard dynamic range (SDR) image compression. However, due to the wide luminance distribution of high dynamic range (HDR) images and the lack of large standard datasets, developing a deep model for HDR image compression is much more challenging. To tackle this issue, we view HDR data as distributional shifts of SDR data and the HDR image compression can be modeled as an out-of-distribution generalization (OoD) problem. Herein, we propose a novel out-of-distribution (OoD) HDR image compression framework (OoDHDR-codec). It learns the general representation across HDR and SDR environments, and allows the model to be trained effectively using a large set of SDR datases supplemented with much fewer HDR samples. Specifically, OoDHDR-codec consists of two branches to process the data from two environments. The SDR branch is a standard blackbox network. For the HDR branch, we develop a hybrid system that models luminance masking and tone mapping with white-box modules and performs content compression with black-box neural networks. To improve the generalization from SDR training data on HDR data, we introduce an invariance regularization term to learn the common representation for both SDR and HDR compression. Extensive experimental results show that the OoDHDR codec achieves strong competitive in-distribution performance and state-of-the-art OoD performance. To the best of our knowledge, our proposed approach is the first work to model HDR compression as OoD generalization problems and our OoD generalization algorithmic framework can be applied to any deep compression model in addition to the network architectural choice demonstrated in the paper. Code available at https://github.com/caolinfeng/OoDHDR-codec.

Learned Block-Based Hybrid Image Compression

Recent works on learned image compression perform encoding and decoding processes in a full-resolution manner, resulting in two problems when deployed for practical applications. First, parallel acceleration of the autoregressive entropy model cannot be achieved due to serial decoding. Second, full-resolution inference often causes the out-of-memory (OOM) problem with limited GPU resources, especially for high-resolution images. Block partition is a good choice to handle the above issues, but it brings about new challenges in reducing the redundancy between blocks and eliminating block effects. To tackle the above challenges, this paper provides a learned block-based hybrid image compression (LBHIC) framework. Specifically, we introduce explicit intra prediction into a learned image compression framework to utilize the relation among adjacent blocks. Superior to context modeling by linear weighting of neighbor pixels in traditional codecs, we propose a contextual prediction module (CPM) to better capture long-range correlations by utilizing the strip pooling to extract the most relevant information in neighboring latent space, thus achieving effective information prediction. Moreover, to alleviate blocking artifacts, we further propose a boundary-aware postprocessing module (BPM) with the edge importance taken into account. Extensive experiments demonstrate that the proposed LBHIC codec outperforms the VVC, with a bit-rate conservation of 4.1%, and reduces the decoding time by approximately 86.7% compared with that of state-of-the-art learned image compression methods.

An efficient and Highly Scalable Listless SPIHT Image Compression Framework

The set partitioning in hierarchical trees is a powerful image compression algorithm. It has reasonable complexity and produces a rate scalable bit-stream. Unfortunately, SPIHT fails to explore the multi-resolution nature of the wavelet transform as the output bit-stream doesn't support resolution scalability. Moreover it requires huge memory and has complex memory management as it depends on utilizing lists with memory of about 2.5 the image size. This paper proposes three related algorithms. The first one modifies SPIHT to reduce its complexity and improve its efficiency especially at low rates. The second is the main contribution of the paper. It provides a simultaneous solution to the memory and scalability problems of SPIHT. Memory is reduced by utilizing status bits of average 2.5 bits per pixel instead of the lists. Resolution scalability is maintained by encoding the resolution levels in increasing order within each coding pass. Another important attribute of our algorithm is that it has very little increment in complexity in comparison with the original SPIHT algorithm. In contrast, the existing solutions have much more complexity, and/or more memory resources. The third has slightly lower complexity and memory than the second but at the same time, it has slightly lower performance.

Scalable compression for machine and human vision tasks via multi-branch shared module

The existing image compression methods are mainly aimed at human perception tasks instead of machine vision tasks. Rich features learned by the shallow and deep layers of pre-trained visual geometry group (VGG)-net can serve human perception and machine vision tasks, respectively. To improve the machine analysis capabilities of the human-targeted compression method, we propose a scalable image compression framework at low bit-rates. Specifically, the scalable compression framework is composed of a base layer (BL) and an enhancement layer (EL), which utilize the correlation between the above features to perform machine analysis and human perception tasks, respectively. For effectively utilizing the above two types of features, a multi-branch shared module that utilizes the complementarity of multi-branch convolution kernels to retain compact information for the above features to support BL and EL tasks is proposed. In addition, for further improving the accuracy of machine analysis, the machine-vision importance map is introduced in BL; it adaptively utilizes the spatial and channel information from the deepest layer of VGG-net to guide local bit allocation. When the bit-rate is limited to 0.2 bpp, the average recognition accuracy (Top-1) of the BL of proposed method is 5.2%, 13.4%, 6.0%, and 7.4% higher than that of BPG, Webp, Mentzer, and NIC, respectively, on the ILSVRC2012 verification dataset. Meanwhile, the EL provides good visual experience.

Compressive Sampling of Color Retinal Image Using Spread Spectrum Fourier Sampling and Total Variant

In medical imaging, the application of retinal images demands a lot of retinal photos to analyze and requires efficient compression techniques for retinal image storage. Retinal images must meet stringent quality requirements for clinical data to be accurate and dependable. This paper proposes a compressive sampling (CS) framework for color retinal image (CRI) compression, which relies on spread spectrum Fourier sampling (SSFS) and total variant (TV)-based reconstruction method with three loops of RGB color space, referred to as RGB-TV. In CS, two procedures are performed, i.e., compression and CS reconstruction. In compression steps, SFFS is performed to get a compressed signal from the original CRI with a high compression ratio (CR). While in CS reconstruction, TV-norm and TV proximal operator are exploited for problem optimization to recover original CRI from a compressed signal. In addition, signal-to-noise ratio (SNR), structural similarity (SSIM), and reconstruction time are investigated for the performance metrics of the proposed RGB-TV. The computer simulation result shows that the proposed RGB-TV with a set of CRI of size 512 by 512 pixels can compress until CR = 10 which obtains mean SNR of 22 dB, SSIM 0.84, and reconstruction time of 2.2 seconds.

Enhanced Standard Compatible Image Compression Framework Based on Auxiliary Codec Networks.

Recent deep neural network-based research to enhance image compression performance can be divided into three categories: learnable codecs, postprocessing networks, and compact representation networks. The learnable codec has been designed for end-to-end learning beyond the conventional compression modules. The postprocessing network increases the quality of decoded images using example-based learning. The compact representation network is learned to reduce the capacity of an input image, reducing the bit rate while maintaining the quality of the decoded image. However, these approaches are not compatible with existing codecs or are not optimal for increasing coding efficiency. Specifically, it is difficult to achieve optimal learning in previous studies using a compact representation network due to the inaccurate consideration of the codecs. In this paper, we propose a novel standard compatible image compression framework based on auxiliary codec networks (ACNs). In addition, ACNs are designed to imitate image degradation operations of the existing codec, which delivers more accurate gradients to the compact representation network. Therefore, compact representation and postprocessing networks can be learned effectively and optimally. We demonstrate that the proposed framework based on the JPEG and High Efficiency Video Coding standard substantially outperforms existing image compression algorithms in a standard compatible manner.

Image Compression Using Stochastic-AFD Based Multisignal Sparse Representation.

Adaptive Fourier decomposition (AFD) is a newly developed signal processing tool that can adaptively decompose any single signal using a Szegö kernel dictionary. To process multiple signals, a novel stochastic-AFD (SAFD) theory was recently proposed. The innovation of this study is twofold. First, a SAFD-based general multi-signal sparse representation learning algorithm is designed and implemented for the first time in the literature, which can be used in many signal and image processing areas. Second, a novel SAFD based image compression framework is proposed. The algorithm design and implementation of the SAFD theory and image compression methods are presented in detail. The proposed compression methods are compared with 13 other state-of-the-art compression methods, including JPEG, JPEG2000, BPG, and other popular deep learning-based methods. The experimental results show that our methods achieve the best balanced performance. The proposed methods are based on single image adaptive sparse representation learning, and they require no pre-training. In addition, the decompression quality or compression efficiency can be easily adjusted by a single parameter, that is, the decomposition level. Our method is supported by a solid mathematical foundation, which has the potential to become a new core technology in image compression.

Collaborative image compression and classification with multi-task learning for visual Internet of Things

Widespread deployment of the Internet of Things (IoT) has changed the way that network services are developed, deployed, and operated. Most onboard advanced IoT devices are equipped with visual sensors that form the so-called visual IoT. Typically, the sender would compress images, and then through the communication network, the receiver would decode images, and then analyze the images for applications. However, image compression and semantic inference are generally conducted separately, and thus, current compression algorithms cannot be transplanted for the use of semantic inference directly. A collaborative image compression and classification framework for visual IoT applications is proposed, which combines image compression with semantic inference by using multi-task learning. In particular, the multi-task Generative Adversarial Networks (GANs) are described, which include encoder, quantizer, generator, discriminator, and classifier to conduct simultaneously image compression and classification. The key to the proposed framework is the quantized latent representation used for compression and classification. GANs with perceptual quality can achieve low bitrate compression and reduce the amount of data transmitted. In addition, the design in which two tasks share the same feature can greatly reduce computing resources, which is especially applicable for environments with extremely limited resources. Using extensive experiments, the collaborative compression and classification framework is effective and useful for visual IoT applications.

A lossless compression method for multi-component medical images based on big data mining

In disease diagnosis, medical image plays an important part. Its lossless compression is pretty critical, which directly determines the requirement of local storage space and communication bandwidth of remote medical systems, so as to help the diagnosis and treatment of patients. There are two extraordinary properties related to medical images: lossless and similarity. How to take advantage of these two properties to reduce the information needed to represent an image is the key point of compression. In this paper, we employ the big data mining to set up the image codebook. That is, to find the basic components of images. We propose a soft compression algorithm for multi-component medical images, which can exactly reflect the fundamental structure of images. A general representation framework for image compression is also put forward and the results indicate that our developed soft compression algorithm can outperform the popular benchmarks PNG and JPEG2000 in terms of compression ratio.

Novel hybrid framework for image compression for supportive hardware design of boosting compression

Performing the image compression over the resource constrained hardware is quite a challenging task. Although, there has been various approaches being carried out towards image compression considering the hardware aspect of it, but still there are problems associated with the memory acceleration associated with the entire operation that downgrade the performance of the hardware device. Therefore, the proposed approach presents a cost effective image compression mechanism which offers lossless compression using a unique combination of the non-linear filtering, segmentation, contour detection, followed by the optimization. The compression mechanism adapts analytical approach for significant image compression. The execution of the compression mechanism yields faster response time, reduced mean square error, improved signal quality and significant compression ratio performance.

Efficiently compressing 3D medical images for teleinterventions via CNNs and anisotropic diffusion.

Efficient compression of images while preserving image quality has the potential to be a major enabler of effective remote clinical diagnosis and treatment, since poor Internet connection conditions are often the primary constraint in such services. This paper presents a framework for organ-specific image compression for teleinterventions based on a deep learning approach and anisotropic diffusion filter. The proposed method, deep learning and anisotropic diffusion (DLAD), uses a convolutional neural network architecture to extract a probability map for the organ of interest; this probability map guides an anisotropic diffusion filter that smooths the image except at the location of the organ of interest. Subsequently, a compression method, such as BZ2 and HEVC-visually lossless, is applied to compress the image. We demonstrate the proposed method on three-dimensional (3D) CT images acquired for radio frequency ablation (RFA) of liver lesions. We quantitatively evaluate the proposed method on 151 CT images using peak-signal-to-noise ratio ( ), structural similarity ( ), and compression ratio ( ) metrics. Finally, we compare the assessments of two radiologists on the liver lesion detection and the liver lesion center annotation using 33 sets of the original images and the compressed images. The results show that the method can significantly improve of most well-known compression methods. DLAD combined with HEVC-visually lossless achieves the highest average of 6.45, which is 36% higher than that of the original HEVC and outperforms other state-of-the-art lossless medical image compression methods. The means of and are 70dB and 0.95, respectively. In addition, the compression effects do not statistically significantly affect the assessments of the radiologists on the liver lesion detection and the lesion center annotation. We thus conclude that the method has a high potential to be applied in teleintervention applications.

End-to-end multispectral image compression framework based on adaptive multiscale feature extraction

Multispectral image compression can considerably reduce the volume of data and promote their application. However, conventional single-scale compression schemes, such as JPEG2000 and three-dimensional set partitioning in hierarchical tree (3D-SPIHT), do not accurately preserve the features of images due to the complex features of multispectral images. A compression framework based on adaptive multiscale feature extraction with a convolutional neural network is proposed. First, an adaptive multiscale feature extraction module, which is the basic component of the compression framework, is designed to extract the multiscale spatial–spectral features of the multispectral images and adaptively adjust the weights of the features according to the content of the images. Second, the encoder, which is composed of multiscale feature extraction modules, extracts the multiscale spatial–spectral features of the multispectral images, and the extracted features are quantized and encoded by the quantizer and the entropy coder to generate a compressed bitstream. Third, the decoder, which is structurally similar to the encoder, is utilized to recover the images. The rate-distortion optimizer is embedded in the encoder to control the trade-off between the rate loss and the distortion. The results of these experiments on multispectral images of the Landsat 8 satellite and the WorldView-3 satellite validate the better performance of our compression framework compared with the performances of conventional schemes, including JPEG2000 and 3D-SPIHT. In order to further verify the effectiveness of multiscale features, the framework is compared with a single-scale compression algorithm based on deep learning, the experimental results validate that the performance of the single-scale compression algorithm superior to the conventional schemes but inferior to our multiscale algorithm, which indicates that the multiscale features can significantly improve the performance of the compression algorithm.

Image compression based on Gaussian mixture model constrained using Markov random field

We introduce a Gaussian Mixture Model (GMM) constrained by Markov Random Field (MRF) framework for image compression in this paper. The image is predicted using GMM with MRF and the parameters of the GMM are estimated using an adjusted Expectation-Maximization (EM) algorithm. Mixture Model Optimization (MMO) is used in this framework to select the optimal number of distributions and avoid local optimum of EM at the same time. Parameters are encoded using fixed-length bits. A codebook is used to improve the coding efficiency of the covariance parameters. The residual between the original image and the prediction is encoded using High Efficiency Video Coding (HEVC) intra coding. Experimental results show that our method performs better than our previous work, HEVC, JPEG 2000 and Better Portable Graphics (BPG) which is an improved version of HEVC.

Semantic Perceptual Image Compression With a Laplacian Pyramid of Convolutional Networks.

The existing image compression methods usually choose or optimize low-level representation manually. Actually, these methods struggle for the texture restoration at low bit rates. Recently, deep neural network (DNN)-based image compression methods have achieved impressive results. To achieve better perceptual quality, generative models are widely used, especially generative adversarial networks (GAN). However, training GAN is intractable, especially for high-resolution images, with the challenges of unconvincing reconstructions and unstable training. To overcome these problems, we propose a novel DNN-based image compression framework in this paper. The key point is decomposing an image into multi-scale sub-images using the proposed Laplacian pyramid based multi-scale networks. For each pyramid scale, we train a specific DNN to exploit the compressive representation. Meanwhile, each scale is optimized with different aspects, including pixel, semantics, distribution and entropy, for a good "rate-distortion-perception" trade-off. By independently optimizing each pyramid scale, we make each stage manageable and make each sub-image plausible. Experimental results demonstrate that our method achieves state-of-the-art performance, with advantages over existing methods in providing improved visual quality. Additionally, a better performance in the down-stream visual analysis tasks which are conducted on the reconstructed images, validates the excellent semantics-preserving ability of the proposed method.

Learned Multi-Resolution Variable-Rate Image Compression With Octave-Based Residual Blocks

Recently deep learning-based image compression has shown the potential to outperform traditional codecs. However, most existing methods train multiple networks for multiple bit rates, which increase the implementation complexity. In this paper, we propose a new variable-rate image compression framework, which employs generalized octave convolutions (GoConv) and generalized octave transposed-convolutions (GoTConv) with built-in generalized divisive normalization (GDN) and inverse GDN (IGDN) layers. Novel GoConv- and GoTConv-based residual blocks are also developed in the encoder and decoder networks. Our scheme also uses a stochastic rounding-based scalar quantization. To further improve the performance, we encode the residual between the input and the reconstructed image from the decoder network as an enhancement layer. To enable a single model to operate with different bit rates and to learn multi-rate image features, a new objective function is introduced. Experimental results show that the proposed framework trained with variable-rate objective function outperforms the standard codecs such as H.265/HEVC-based BPG and state-of-the-art learning-based variable-rate methods.

Spectral–Spatial Feature Partitioned Extraction Based on CNN for Multispectral Image Compression

Recently, the rapid development of multispectral imaging technology has received great attention from many fields, which inevitably involves the image transmission and storage problem. To solve this issue, a novel end-to-end multispectral image compression method based on spectral–spatial feature partitioned extraction is proposed. The whole multispectral image compression framework is based on a convolutional neural network (CNN), whose innovation lies in the feature extraction module that is divided into two parallel parts, one is for spectral and the other is for spatial. Firstly, the spectral feature extraction module is used to extract spectral features independently, and the spatial feature extraction module is operated to obtain the separated spatial features. After feature extraction, the spectral and spatial features are fused element-by-element, followed by downsampling, which can reduce the size of the feature maps. Then, the data are converted to bit-stream through quantization and lossless entropy encoding. To make the data more compact, a rate-distortion optimizer is added to the network. The decoder is a relatively inverse process of the encoder. For comparison, the proposed method is tested along with JPEG2000, 3D-SPIHT and ResConv, another CNN-based algorithm on datasets from Landsat-8 and WorldView-3 satellites. The result shows the proposed algorithm outperforms other methods at the same bit rate.

Perceptual Image Compression with Block-Level Just Noticeable Difference Prediction

A block-level perceptual image compression framework is proposed in this work, including a block-level just noticeable difference (JND) prediction model and a preprocessing scheme. Specifically speaking, block-level JND values are first deduced by utilizing the OTSU method based on the variation of block-level structural similarity values between two adjacent picture-level JND values in the MCL-JCI dataset. After the JND value for each image block is generated, a convolutional neural network–based prediction model is designed to forecast block-level JND values for a given target image. Then, a preprocessing scheme is devised to modify the discrete cosine transform coefficients during JPEG compression on the basis of the distribution of block-level JND values of the target test image. Finally, the test image is compressed by the max JND value across all of its image blocks in the light of the initial quality factor setting. The experimental results demonstrate that the proposed block-level perceptual image compression method is able to achieve 16.75% bit saving as compared to the state-of-the-art method with similar subjective quality. The project page can be found at https://mic.tongji.edu.cn/43/3f/c9778a148287/page.htm.

MAGIC: Machine-Learning-Guided Image Compression for Vision Applications in Internet of Things

The emergent ecosystems of intelligent edge devices in diverse Internet-of-Things (IoT) applications, from automatic surveillance to precision agriculture, increasingly rely on recording and processing a variety of image data. Due to resource constraints, e.g., energy and communication bandwidth requirements, these applications require compressing the recorded images before transmission. For these applications, image compression commonly requires: 1) maintaining features for coarse-grain pattern recognition instead of the high-level details for human perception due to machine-to-machine communications; 2) high compression ratio that leads to improved energy and transmission efficiency; and 3) large dynamic range of compression and an easy tradeoff between compression factor and quality of reconstruction to accommodate a wide diversity of IoT applications as well as their time-varying energy/performance needs. To address these requirements, we propose, MAGIC, a novel machine learning (ML)-guided image compression framework that judiciously sacrifices the visual quality to achieve much higher compression when compared to traditional techniques, while maintaining accuracy for coarse-grained vision tasks. The central idea is to capture application-specific domain knowledge and efficiently utilize it in achieving high compression. We demonstrate that the MAGIC framework is configurable across a wide range of compression/quality and is capable of compressing beyond the standard quality factor limits of both JPEG 2000 and WebP. We perform experiments on representative IoT applications using two vision data sets and show 42.65× compression at similar accuracy with respect to the source. We highlight low variance in compression rate across images using our technique as compared to JPEG 2000 and WebP.

Achieving a reversible lower dimensionality transformation for picture archiving and communication system in healthcare

With the progression of picture archiving and communication systems (PACSs) over the past decade, it has become imperative that such systems be optimised in security, storage, and transmission aspects. The work presented in this Letter shows a framework for medical image compression and secure image transmission for PACSs. The work aims to achieve a lower dimensionality of input medical image signified by a high-compression ratio, a secure image transmission that can withstand adversarial attacks and provide a reversible reconstruction with minimal error. The authors illustrate that sinusoid modulated Gaussian texture maps, multi-level chaotic maps, and high-frequency image maps can be efficiently fused and utilised in a deep learning architecture. The overall analysis depicts promising results with regard to the capability of image compression, security, and transmission. The proposed framework will be a potential candidate for use in PACSs, which effectively is the backbone of the current healthcare paradigm.

A residual network framework based on weighted feature channels for multispectral image compression

Deep learning has achieved great success in many computer vision tasks, such as recognition, image enhancement and image compression. However, it is difficult to use a residual network that can efficiently consider the characteristics of multispectral images for multispectral image compression. In this paper, a novel end-to-end multispectral image compression framework based on a weighted feature channel residual network is proposed to efficiently remove the spatial and spectral redundancy of multispectral images by extracting the importance of each channel. The multispectral image compression framework includes a forward coding network, a rate-distortion optimizer, a quantizer/inverse quantizer, an entropy encoder/decoder and an inverse decoding network. In the encoder, multispectral images are directly fed into the forward coding network, and the main spectral and spatial features of the multispectral images are extracted by the residual block. Additionally, the weighted feature channel module can explicitly model the relationship between feature channels when extracting features from multispectral images and adaptively allocate different weights for each feature channel through training. The rate-distortion optimizer is added to make the main features compact. Then, the intermediate feature data are quantized and encoded by lossless entropy coding to obtain a code stream. In the decoder, the code stream is approximately restored to the intermediate features through the entropy decoder and the inverse quantizer. Then, the intermediate features are reconstructed to the multispectral images by the inverse decoding network. Experimental results on 7-band multispectral images of the Landsat 8 satellite and 8-band multispectral images of WorldView-3 satellite demonstrate that the proposed algorithm can achieve a better PSNR than conventional 2D schemes (which are JPEG2000 and JPEG in this paper) and 3D scheme (which is 3D-SPIHT in this paper) and can effectively preserve more spectral information of multispectral images.