Scalable Convolutional Neural Network for Image Compressed Sensing

Shi, Wuzhen; Jiang, Feng; Liu, Shaohui; Zhao, Debin

doi:10.1109/cvpr.2019.01257

Cited by 131 publications

(118 citation statements)

References 29 publications

Supporting

Mentioning

118

Contrasting

Order By: Relevance

“…In order to estimate the model parameters of the proposed average codeword length model and the relative PSNR model, 100 images in the BSDS500 dataset [35] were randomly selected for training, and the BSD68 dataset [36] was used for testing, each image being cropped to a 256 × 256 size. During training, the quantization bit-depth took eight values in {3, 4, .…”

Section: Model Parameter Estimation For the Bit-rate Model And The Rementioning

confidence: 99%

Low-Complexity Rate-Distortion Optimization of Sampling Rate and Bit-Depth for Compressed Sensing of Images

Chen

Gong

et al. 2020

Entropy

View full text Add to dashboard Cite

Compressed sensing (CS) offers a framework for image acquisition, which has excellent potential in image sampling and compression applications due to the sub-Nyquist sampling rate and low complexity. In engineering practices, the resulting CS samples are quantized by finite bits for transmission. In circumstances where the bit budget for image transmission is constrained, knowing how to choose the sampling rate and the number of bits per measurement (bit-depth) is essential for the quality of CS reconstruction. In this paper, we first present a bit-rate model that considers the compression performance of CS, quantification, and entropy coder. The bit-rate model reveals the relationship between bit rate, sampling rate, and bit-depth. Then, we propose a relative peak signal-to-noise ratio (PSNR) model for evaluating distortion, which reveals the relationship between relative PSNR, sampling rate, and bit-depth. Finally, the optimal sampling rate and bit-depth are determined based on the rate-distortion (RD) criteria with the bit-rate model and the relative PSNR model. The experimental results show that the actual bit rate obtained by the optimized sampling rate and bit-depth is very close to the target bit rate. Compared with the traditional CS coding method with a fixed sampling rate, the proposed method provides better rate-distortion performance, and the additional calculation amount amounts to less than 1%.

show abstract

Section: Model Parameter Estimation For the Bit-rate Model And The Rementioning

confidence: 99%

Low-Complexity Rate-Distortion Optimization of Sampling Rate and Bit-Depth for Compressed Sensing of Images

Chen

Gong

et al. 2020

Entropy

View full text Add to dashboard Cite

show abstract

“…Specifically, our method outperforms NLR-CS and BM3D-CS by 2.29 dB and 1.95 dB on the Waterloo140 dataset at a sampling ratio of 0.10. We further compare our method with some advanced deepbased CS image reconstruction methods (including ISTA-Net+ [11], CSNet+ [46], SCSNet [15], OPINE-Net+ [47], AMP-Net [13], and LDAMP [18]) on the Set8 and Water-loo140 datasets. Table III provides the PSNR values of the competing methods for every image in Set8, and Table IV lists the average PSNR results of seven classes of Waterloo140.…”

Section: B Comparison With State-of-the-art Methodsmentioning

confidence: 99%

“…1, a model-fixed end-to-end approach cannot reconstruct the image accurately once the measurement loss rate increases to some level. SCSNet [15] achieves a scalable CS end-to-end net by using a greedy strategy to search the most important measurement bases. However, SCSNet still needs to update the network parameters for different sampling ratios, and the complexity of greedy searching is no less than that of retraining the model at a high sampling ratio.…”

Section: Introductionmentioning

confidence: 99%

Deep-Learned Regularization and Proximal Operator for Image Compressive Sensing

Chen

Guo

Feng

et al. 2021

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Deep learning has recently been intensively studied in the context of image compressive sensing (CS) to discover and represent complicated image structures. These approaches, however, either suffer from nonflexibility for an arbitrary sampling ratio or lack an explicit deep-learned regularization term. This paper aims to solve the CS reconstruction problem by combining the deep-learned regularization term and proximal operator. We first introduce a regularization term using a carefully designed residual-regressive net, which can measure the distance between a corrupted image and a clean image set and accurately identify to which subspace the corrupted image belongs. We then address a proximal operator with a tailored dilated residual channel attention net, which enables the learned proximal operator to map the distorted image into the clean image set. We adopt an adaptive proximal selection strategy to embed the network into the loop of the CS image reconstruction algorithm. Moreover, a self-ensemble strategy is presented to improve CS recovery performance. We further utilize state evolution to analyze the effectiveness of the designed networks. Extensive experiments also demonstrate that our method can yield superior accurate reconstruction (PSNR gain over 1 dB) compared to other competing approaches while achieving the current state-of-the-art image CS reconstruction performance. The test code is available at https://github.com/zjut-gwl/CSDRCANet.

show abstract

“…(2) will become inefficient and even impractical if the dimension of the dictionary is high or the size of training dataset is very large [13], [16]. Lately, some deep networks are developed to jointly optimizing the sampling matrix and the non-linear recovery operator [17], [18], [19], [45], [20], [46]. In particular, Adler et al propose to utilize a fully-connected network to perform both the block-based linear sensing and non-linear reconstruction stages.…”

Section: Related Workmentioning

confidence: 99%

Optimization-Inspired Compact Deep Compressive Sensing

Zhang

Zhao

Gao

2020

IEEE J. Sel. Top. Signal Process.

157

View full text Add to dashboard Cite

In order to improve CS performance of natural images, in this paper, we propose a novel framework to design an OPtimization-INspired Explicable deep Network, dubbed OPINE-Net, for adaptive sampling and recovery. Both orthogonal and binary constraints of sampling matrix are incorporated into OPINE-Net simultaneously. In particular, OPINE-Net is composed of three subnets: sampling subnet, initialization subnet and recovery subnet, and all the parameters in OPINE-Net (e.g. sampling matrix, nonlinear transforms, shrinkage threshold) are learned end-to-end, rather than hand-crafted. Moreover, considering the relationship among neighboring blocks, an enhanced version OPINE-Net + is developed, which allows image blocks to be sampled independently but reconstructed jointly to further enhance the performance. In addition, some interesting findings of learned sampling matrix are presented. Compared with existing state-of-the-art network-based CS methods, the proposed hardware-friendly OPINE-Nets not only achieve better performance but also require much fewer parameters and much less storage space, while maintaining a real-time running speed.

show abstract

Scalable Convolutional Neural Network for Image Compressed Sensing

Cited by 131 publications

References 29 publications

Low-Complexity Rate-Distortion Optimization of Sampling Rate and Bit-Depth for Compressed Sensing of Images

Low-Complexity Rate-Distortion Optimization of Sampling Rate and Bit-Depth for Compressed Sensing of Images

Deep-Learned Regularization and Proximal Operator for Image Compressive Sensing

Optimization-Inspired Compact Deep Compressive Sensing

Contact Info

Product

Resources

About