An Efficient Compressive Sensed Video Codec with Inter-Frame Decoding and Low-Complexity Intra-Frame Encoding

Belyaev, Evgeny

doi:10.3390/s23031368

Cited by 6 publications

(6 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We proposed an orthogonal complex-valued Gaussian measurement matrix constructed by using two Gaussian matrices after Gram-Schmidt orthogonalization as the Table 4 shows the addition and multiplication complexity of several different matrices during the measurement process. In references [7,8], the Haar wavelet transform and the Noiselet transform were used to realize compression. For these two cases, the Haar wavelet transform only needs to multiply 1 √ 2…”

Section: Discussionmentioning

confidence: 99%

“…Table 4 shows the addition and multiplication complexity of several different matrices during the measurement process. In references [ 7 , 8 ], the Haar wavelet transform and the Noiselet transform were used to realize compression. For these two cases, the Haar wavelet transform only needs to multiply

, which can be realized via a simple shift operation, and the Noiselet transform only needs addition operations.…”

Section: Experimental Analysismentioning

confidence: 99%

“…By combining the process of sampling and compression, compressed sensing can acquire a signal at a lower sampling rate and accurately recover the original signal, thereby substantially reducing the overhead of data acquisition and storage. After long-term development, compressed sensing has found widespread application in numerous domains, including wireless communication [ 4 ], medical imaging [ 5 ], smart city construction [ 6 ], and video codec [ 7 , 8 ]. Compressed sensing theory primarily consists of three components: the sparse representation of the signal, the measurement matrix’s design, and the reconstruction algorithm’s design.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Novel Complex-Valued Gaussian Measurement Matrix for Image Compressed Sensing

Wang,

Xue,

Yan

et al. 2023

Entropy

View full text Add to dashboard Cite

The measurement matrix used influences the performance of image reconstruction in compressed sensing. To enhance the performance of image reconstruction in compressed sensing, two different Gaussian random matrices were orthogonalized via Gram–Schmidt orthogonalization, respectively. Then, one was used as the real part and the other as the imaginary part to construct a complex-valued Gaussian matrix. Furthermore, we sparsified the proposed measurement matrix to reduce the storage space and computation. The experimental results show that the complex-valued Gaussian matrix after orthogonalization has better image reconstruction performance, and the peak signal-to-noise ratio and structural similarity under different compression ratios are better than the real-valued measurement matrix. Moreover, the sparse measurement matrix can effectively reduce the amount of calculation.

show abstract

Section: Discussionmentioning

confidence: 99%

, which can be realized via a simple shift operation, and the Noiselet transform only needs addition operations.…”

Section: Experimental Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Complex-Valued Gaussian Measurement Matrix for Image Compressed Sensing

Wang,

Xue,

Yan

et al. 2023

Entropy

View full text Add to dashboard Cite

show abstract

“…Note that some advanced sensing approaches have been proposed for a better recovery quality such as Ref. [34], and it is easiy to replace the BCS with these promising methods in our network.…”

Section: Methodsmentioning

confidence: 99%

A temporal shift reconstruction network for compressive video sensing

Gu,

Zhou,

Lin

2023

IET Computer Vision

View full text Add to dashboard Cite

Compressive sensing provides a promising sampling paradigm for video acquisition for resource‐limited sensor applications. However, the reconstruction of original video signals from sub‐sampled measurements is still a great challenge. To exploit the temporal redundancies within videos during the recovery, previous works tend to perform alignment on initial reconstructions, which are too coarse to provide accurate motion estimations. To solve this problem, the authors propose a novel reconstruction network, named TSRN, for compressive video sensing. Specifically, the authors utilise a number of stacked temporal shift reconstruction blocks (TSRBs) to enhance the initial reconstruction progressively. Each TSRB could learn the temporal structures by exchanging information with last and next time step, and no additional computations is imposed on the network compared to regular 2D convolutions due to the high efficiency of temporal shift operations. After the enhancement, a bidirectional alignment module to build accurate temporal dependencies directly with the help of optical flows is employed. Different from previous methods that only extract supplementary information from the key frames, the proposed alignment module can receive temporal information from the whole video sequence via bidirectional propagations, thus yielding better performance. Experimental results verify the superiority of the proposed method over other state‐of‐the‐art approaches quantitatively and qualitatively.

show abstract

“…Reconstruction of the image is usually obtained using algorithms that leverage the sparsity of the signal, such as sparse optimization or convex optimization techniques, but these techniques are often computationally expensive and slow to converge. CS finds applications in various fields, including magnetic resonance imaging [3], radar signal sampling [4], cryptosystems [5], snapshot imaging [6], and video sensing [7,8]. It proves especially useful when dealing with large amounts of data, as it can lead to significant reductions in storage and processing requirements.…”

Section: Introductionmentioning

confidence: 99%

MsDC-DEQ-Net: Deep Equilibrium Model (DEQ) with Multiscale Dilated Convolution for Image Compressive Sensing (CS)

Yu,

Dansereau

2024

IET Signal Processing

View full text Add to dashboard Cite

Compressive sensing (CS) is a technique that enables the recovery of sparse signals using fewer measurements than traditional sampling methods. To address the computational challenges of CS reconstruction, our objective is to develop an interpretable and concise neural network model for reconstructing natural images using CS. We achieve this by mapping one step of the iterative shrinkage thresholding algorithm (ISTA) to a deep network block, representing one iteration of ISTA. To enhance learning ability and incorporate structural diversity, we integrate aggregated residual transformations (ResNeXt) and squeeze-and-excitation mechanisms into the ISTA block. This block serves as a deep equilibrium layer connected to a semi-tensor product network for convenient sampling and providing an initial reconstruction. The resulting model, called MsDC-DEQ-Net, exhibits competitive performance compared to state-of-the-art network-based methods. It significantly reduces storage requirements compared to deep unrolling methods, using only one iteration block instead of multiple iterations. Unlike deep unrolling models, MsDC-DEQ-Net can be iteratively used, gradually improving reconstruction accuracy while considering computation tradeoffs. Additionally, the model benefits from multiscale dilated convolutions, further enhancing performance.

show abstract

An Efficient Compressive Sensed Video Codec with Inter-Frame Decoding and Low-Complexity Intra-Frame Encoding

Cited by 6 publications

References 31 publications

A Novel Complex-Valued Gaussian Measurement Matrix for Image Compressed Sensing

A Novel Complex-Valued Gaussian Measurement Matrix for Image Compressed Sensing

A temporal shift reconstruction network for compressive video sensing

MsDC-DEQ-Net: Deep Equilibrium Model (DEQ) with Multiscale Dilated Convolution for Image Compressive Sensing (CS)

Contact Info

Product

Resources

About