scispace - formally typeset
Search or ask a question
Topic

Upsampling

About: Upsampling is a research topic. Over the lifetime, 2426 publications have been published within this topic receiving 57613 citations.


Papers
More filters
Patent
17 Nov 2017
TL;DR: Zhang et al. as mentioned in this paper proposed a single-image super-resolution reconstruction method based on a deep residual network, which mainly comprises a first step of performing block extraction and pixel averaging processing on an image in a sample image database to obtain a corresponding high-resolution and low-resolution training image sets; a second step of constructing a deep convolutional neutral network with a residual structure for iterative training, and then inputting the training set obtained in the first step to the neural network constructed in the second step, according to a data model obtained by training, realizing the
Abstract: The invention discloses a single-image super-resolution reconstruction method based on a deep residual network. The method of the invention mainly comprises a first step of performing block extraction and pixel averaging processing on an image in a sample image database to obtain a corresponding high resolution and low resolution training image sets; a second step of constructing a deep convolutional neutral network with a residual structure for iterative training, and then inputting the training set obtained in the first step to the neural network constructed in the second step for iterative training; and a third step of according to a data model obtained by training, realizing the continuous up-scaling of the input low resolution image through the combination of iterative operation and an interpolation algorithm. By introducing a deep residual network and introducing an upsampling layer at the end of the network, the method of the invention accelerates the processing speed of the image up-scaling, enhances the display effect of the image details, obtains a better image super-resolution reconstruction effect, and has a wide range of applications in the image high definition display, image compression, security checks and other fields.

13 citations

Journal ArticleDOI
TL;DR: In this paper , an improved lightweight algorithm based on You Only Look Once (YOLOv5) is proposed, which screens for the multiscale detection structure that is suitable for the flat jujube by adjusting the number of layers of target detection, which improves the accuracy of detection and reduces the nuisance parameter.

13 citations

Patent
04 Jun 1984
TL;DR: In this article, signal samples from two signal streams are downsampled, filtered and multiplexed for transmission by being fed in parallel to a switching network which feeds samples from the two signal stream (X, Y) alternately to one or other of two phases of filter.
Abstract: Signal samples from two signal streams are downsampled, filtered and multiplexed for transmission by being fed in parallel to a switching network which feeds samples from the two signal streams (X, Y) alternately to one or other of two phases of filter. Each phase applies a plurality of coefficients (15) to each received sample and then sums the samples to provide a resultant downsampled and multiplexed signal. Receiving apparatus carries out the reverse operation and with interpolation the two original signal streams are reconstructed.

13 citations

Posted Content
TL;DR: In this article, a pixel-adaptive convolution (PAC) operation is proposed, in which the filter weights are multiplied with a spatially-varying kernel that depends on learnable, local pixel features.
Abstract: Convolutions are the fundamental building block of CNNs. The fact that their weights are spatially shared is one of the main reasons for their widespread use, but it also is a major limitation, as it makes convolutions content agnostic. We propose a pixel-adaptive convolution (PAC) operation, a simple yet effective modification of standard convolutions, in which the filter weights are multiplied with a spatially-varying kernel that depends on learnable, local pixel features. PAC is a generalization of several popular filtering techniques and thus can be used for a wide range of use cases. Specifically, we demonstrate state-of-the-art performance when PAC is used for deep joint image upsampling. PAC also offers an effective alternative to fully-connected CRF (Full-CRF), called PAC-CRF, which performs competitively, while being considerably faster. In addition, we also demonstrate that PAC can be used as a drop-in replacement for convolution layers in pre-trained networks, resulting in consistent performance improvements.

13 citations

Proceedings ArticleDOI
F. Baumgarte1
07 May 2001
TL;DR: A suitable analysis filter-bank structure employing cascaded low-order IIR filters and appropriate downsampling to increase efficiency is presented, which enables improved masking modeling for audio coding at low computational costs.
Abstract: Many applications in auditory modeling require analysis filters that approximate the frequency selectivity given by psychophysical data, e.g., from masking experiments using narrow-band maskers. This frequency selectivity is largely determined by the spectral decomposition process inside the human cochlea. Currently used spectral decomposition schemes for masking modeling in audio coding generally do not achieve the non-uniform time and frequency resolution provided by the cochlea. These applications rather take advantage of the computational efficiency of uniform filter banks or transforms at the expense of coding gain. This paper presents a suitable analysis filter-bank structure employing cascaded low-order IIR filters and appropriate downsampling to increase efficiency. In an application example, the filter responses were optimized to model auditory masking effects. The results show that the time and frequency resolution of the filter bank matches or exceeds the masking properties. Thus, the filter bank enables improved masking modeling for audio coding at low computational costs.

13 citations


Network Information
Related Topics (5)
Convolutional neural network
74.7K papers, 2M citations
90% related
Image segmentation
79.6K papers, 1.8M citations
90% related
Feature extraction
111.8K papers, 2.1M citations
89% related
Deep learning
79.8K papers, 2.1M citations
88% related
Feature (computer vision)
128.2K papers, 1.7M citations
87% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
2023469
2022859
2021330
2020322
2019298
2018236