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
Proceedings ArticleDOI
14 Jun 2020
TL;DR: This work introduces a method for creating an accurate sky-mask dataset that is based on partially annotated images that are inpainted and refined by the authors' modified weighted guided filter, and uses this dataset to train a neural network for semantic sky segmentation.
Abstract: The sky is a major component of the appearance of a photograph, and its color and tone can strongly influence the mood of a picture. In nighttime photography, the sky can also suffer from noise and color artifacts. For this reason, there is a strong desire to process the sky in isolation from the rest of the scene to achieve an optimal look. In this work, we propose an automated method, which can run as a part of a camera pipeline, for creating accurate sky alpha-masks and using them to improve the appearance of the sky. Our method performs end-to-end sky optimization in less than half a second per image on a mobile device. We introduce a method for creating an accurate sky-mask dataset that is based on partially annotated images that are inpainted and refined by our modified weighted guided filter. We use this dataset to train a neural network for semantic sky segmentation. Due to the compute and power constraints of mobile devices, sky segmentation is performed at a low image resolution. Our modified weighted guided filter is used for edge-aware upsampling to resize the alpha-mask to a higher resolution. With this detailed mask we automatically apply post-processing steps to the sky in isolation, such as automatic spatially varying white-balance, brightness adjustments, contrast enhancement, and noise reduction.

10 citations

Proceedings ArticleDOI
30 Aug 2021
TL;DR: NU-Wave as mentioned in this paper is the first neural audio upsampling model to produce waveforms of sampling rate 48kHz from coarse 16kHz or 24kHz inputs, while prior works could generate only up to 16kHz.
Abstract: In this work, we introduce NU-Wave, the first neural audio upsampling model to produce waveforms of sampling rate 48kHz from coarse 16kHz or 24kHz inputs, while prior works could generate only up to 16kHz. NU-Wave is the first diffusion probabilistic model for audio super-resolution which is engineered based on neural vocoders. NU-Wave generates high-quality audio that achieves high performance in terms of signal-to-noise ratio (SNR), log-spectral distance (LSD), and accuracy of the ABX test. In all cases, NU-Wave outperforms the baseline models despite the substantially smaller model capacity (3.0M parameters) than baselines (5.4-21%). The audio samples of our model are available at this https URL, and the code will be made available soon.

10 citations

Patent
10 Apr 2014
TL;DR: In this article, a sampling filter process for scalable video coding is provided for re-sampling using video data obtained from an encoder or decoder process of a base layer (BL) in a multi-layer system using adaptive phase shifting to improve quality in Scalable High efficiency Video Coding (SHVC).
Abstract: A sampling filter process is provided for scalable video coding. The process provides for re-sampling using video data obtained from an encoder or decoder process of a base layer (BL) in a multi-layer system using adaptive phase shifting to improve quality in Scalable High efficiency Video Coding (SHVC). In order to compensate for phase offsets introduced by downsampling an appropriate phase offset adjustment is made for upsampling in SHVC with an appropriate offset included for proper luma/chroma color space positions. In one approach the luma/chroma phase offset is specified and a filter is selected to apply the appropriate phase change.

10 citations

Proceedings ArticleDOI
Jinlin Ma1, Meng Wei1, Ziping Ma1, Li Shi1, Kai Zhu1 
01 Aug 2019
TL;DR: This paper proposes a method for combining dilated convolutions with U-Net networks to detect small vessels by using the least squares loss function to replace the Sigmoid cross entropy, which alleviates the problem of gradient disappearance.
Abstract: Retinal vessel segmentation, which can obtain rich eye information, is an important indicator for examining ophthalmology and cardiovascular diseases. The existing method results in a decrease in resolution due to continuous maximum pooling and downsampling operations, causing the network to lose detail on the image when upsampling and restoring the feature map, ignoring small vessels or allow false positives at terminal branches. In order to better solve these problems, this paper proposes a method for combining dilated convolutions with U-Net networks to detect small vessels. In order to force the generated samples to fit the real sample distribution as much as possible, this paper uses the least squares loss function to replace the Sigmoid cross entropy, which alleviates the problem of gradient disappearance. The experimental results show that the proposed method improves the segmentation accuracy and reduces the loss to a greater extent on the DRIVE and STARE datasets.

10 citations

Journal ArticleDOI
TL;DR: A deep convolutional autoencoder neural network for denoising, which consists of encoding and decoding frameworks that can suppress random noise effectively and is tested on synthetic and field cases.

10 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