26,193 Hits in 6.0 sec

Super-resolved multi-temporal segmentation with deep permutation-invariant networks [article]

Diego Valsesia, Enrico Magli
2022 arXiv   pre-print
Multi-image super-resolution from multi-temporal satellite acquisitions of a scene has recently enjoyed great success thanks to new deep learning models.  ...  In this paper, we go beyond classic image reconstruction at a higher resolution by studying a super-resolved inference problem, namely semantic segmentation at a spatial resolution higher than the one  ...  This serves as an enhancement for the segmentation task, with respect to pure super-resolution, because it allows features to exploit a more global context, rather than localized properties.  ... 
arXiv:2204.02631v1 fatcat:wc4ngkj4fzbalnyvpnjo2vofwq

Deep learning approaches for real-time image super-resolution

Pourya Shamsolmoali, M. Emre Celebi, Ruili Wang
2020 Neural computing & applications (Print)  
Generating a high-resolution (HR) image from its corresponding low-resolution (LR) input is referred to image super-resolution (SR).  ...  The paper entitled "Perceptual image quality using dual generative adversarial network" develops a variety of generative adversarial networks for image SR that contains two generators and two discriminators  ...  In ''CASR: a context-aware residual network for singleimage super-resolution,'' the authors propose a lightweight context-aware deep residual network, which appropriately encodes channel and spatial attention  ... 
doi:10.1007/s00521-020-05176-z fatcat:gvmbzve6kvfmzfwumblipqzaji

Improved Super Resolution of MR Images Using CNNs and Vision Transformers [article]

Dwarikanath Mahapatra
2022 arXiv   pre-print
We combine local information of CNNs and global information from ViTs for image super resolution and output super resolved images that have superior quality than those produced by state of the art methods  ...  State of the art magnetic resonance (MR) image super-resolution methods (ISR) using convolutional neural networks (CNNs) leverage limited contextual information due to the limited spatial coverage of CNNs  ...  For the actual super resolution at inference stage it took 1.2 seconds for 2x upsampling for a 512 × 512 image E Architecture of Super Resolution Network Figure 6 shows the detailed architecture of  ... 
arXiv:2207.11748v1 fatcat:25u4xyvmu5agraugfyes57xi54

S2A: Wasserstein GAN with Spatio-Spectral Laplacian Attention for Multi-Spectral Band Synthesis [article]

Litu Rout, Indranil Misra, S Manthira Moorthi, Debajyoti Dhar
2020 arXiv   pre-print
In this regard, we introduce a new cost function for the discriminator based on spatial attention and domain adaptation loss.  ...  Intersection of adversarial learning and satellite image processing is an emerging field in remote sensing.  ...  For this reason, we use upsampled coarse resolution image to compute the attention maps.  ... 
arXiv:2004.03867v1 fatcat:6n7jdgf7r5crnkrkqa4ulnt5ya

Multi-image Super Resolution of Remotely Sensed Images using Residual Feature Attention Deep Neural Networks [article]

Francesco Salvetti, Vittorio Mazzia, Aleem Khaliq, Marcello Chiaberge
2020 arXiv   pre-print
state-of-the-art for Multi-Image Super-Resolution for remote sensing applications.  ...  In this context, the presented research proposes a novel residual attention model (RAMS) that efficiently tackles the multi-image super-resolution task, simultaneously exploiting spatial and temporal correlations  ...  [30] proposed a novel multi-perception attention network (MPSR) for Super-resolution of low resolution remotely sensed images, which achieved better results by incorporating the proposed enhanced residual  ... 
arXiv:2007.03107v2 fatcat:cdyx7nxxnnaxhczwoxggdsjdkm

Multi-Image Super Resolution of Remotely Sensed Images Using Residual Attention Deep Neural Networks

Francesco Salvetti, Vittorio Mazzia, Aleem Khaliq, Marcello Chiaberge
2020 Remote Sensing  
state-of-the-art for Multi-image Super-resolution for remote sensing applications.  ...  In this context, the presented research proposes a novel residual attention model (RAMS) that efficiently tackles the Multi-image Super-resolution task, simultaneously exploiting spatial and temporal correlations  ...  Acknowledgments: This work has been developed with the contribution of the Politecnico di Torino Interdepartmental Centre for Service Robotics PIC4SeR ( and SmartData@Polito (  ... 
doi:10.3390/rs12142207 fatcat:737vstox6bb4vp7gcxydjkavlu

Detail-Preserving Transformer for Light Field Image Super-resolution

Shunzhou Wang, Tianfei Zhou, Yao Lu, Huijun Di
Recently, numerous algorithms have been developed to tackle the problem of light field super-resolution (LFSR), i.e., super-resolving low-resolution light fields to gain high-resolution views.  ...  locally-enhanced self-attention layer, which maintains the locality of each sub-aperture image as well.  ...  Transformer for Image Super-Resolution.  ... 
doi:10.1609/aaai.v36i3.20153 fatcat:tbpf764je5hdlawyndbdqdhaou

Lightweight Non-Local Network for Image Super-Resolution

Risheng Wang, Tao Lei, Wenzheng Zhou, Qi Wang, Hongying Meng, Asoke K. Nandi
2021 ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
The popular deep convolutional networks used for image super-resolution (SR) reconstruction often increase the network depth and employ attention mechanism to improve image reconstruction effect.  ...  To address these issues, we propose a lightweight non-local network (LNLN) for image super resolution in this paper. The proposed network makes two contributions.  ...  Therefore, it is not practical to improve image super-resolution by changing hardware equipment. Due to this reason, researchers begin to pay attention to image super-resolution algorithms.  ... 
doi:10.1109/icassp39728.2021.9414527 fatcat:vrikwd6zhbenljdckh5atau4na

2021 Index IEEE Transactions on Image Processing Vol. 30

2021 IEEE Transactions on Image Processing  
The Author Index contains the primary entry for each item, listed under the first author's name.  ...  ., +, TIP 2021 2908-2922 Interpretable Detail-Fidelity Attention Network for Single Image Super-Resolution.  ...  ., +, TIP 2021 6906-6916 SRGAT: Single Image Super-Resolution With Graph Attention Network.  ... 
doi:10.1109/tip.2022.3142569 fatcat:z26yhwuecbgrnb2czhwjlf73qu

Single Image Joint Motion Deblurring and Super-Resolution Using the Multi-Scale Channel Attention Modules

Misak T. Shoyan, National Polytechnic University of Armenia
2021 Mathematical Problems of Computer Science  
During the last decade, deep convolutional neural networks have significantly advanced the single image super-resolution techniques reconstructing realistic textural and spatial details.  ...  This work proposes a fully convolutional neural network to reconstruct high-resolution sharp images from the given motion blurry low-resolution images.  ...  Introduction Single image super-resolution (SISR) addresses the problem of recovering a sharp, highresolution (HR) image from a given low-resolution (LR) image.  ... 
doi:10.51408/1963-0076 fatcat:ulnois4kvzggvclqywynfcfcfy

LSTNet: A Reference-Based Learning Spectral Transformer Network for Spectral Super-Resolution

Debao Yuan, Ling Wu, Huinan Jiang, Bingrui Zhang, Jian Li
2022 Sensors  
Spectral super-resolution (SSR) is a method that involves learning the relationship between a multispectral image (MSI) and an HSI, based on the overlap region, followed by reconstruction of the HSI by  ...  a reference HSI to create a reasonable reconstruction spectrum.  ...  [40] proposed a texture transformer network for image super-resolution, in which the spatial attention between a low-resolution image and a reference low-resolution image is calculated to further transfer  ... 
doi:10.3390/s22051978 pmid:35271131 pmcid:PMC8914896 fatcat:sqgidsrk4bhqvfmbmlswgicrja

Rethinking Super-Resolution as Text-Guided Details Generation [article]

Chenxi Ma, Bo Yan, Qing Lin, Weimin Tan, Siming Chen
2022 arXiv   pre-print
Deep neural networks have greatly promoted the performance of single image super-resolution (SISR).  ...  To enhance the semantic accuracy and the visual quality of the reconstructed image, we explore the multi-modal fusion learning in SISR by proposing a Text-Guided Super-Resolution (TGSR) framework, which  ...  These explorative SR works reveal more possibilities for image super-resolution.  ... 
arXiv:2207.06604v1 fatcat:s5t3ups4dfhrhiq4czlaml2hla

Implicit LiDAR Network: LiDAR Super-Resolution via Interpolation Weight Prediction [article]

Youngsun Kwon, Minhyuk Sung, Sung-Eui Yoon
2022 arXiv   pre-print
Super-resolution of LiDAR range images is crucial to improving many downstream tasks such as object detection, recognition, and tracking.  ...  While deep learning has made a remarkable advances in super-resolution techniques, typical convolutional architectures limit upscaling factors to specific output resolutions in training.  ...  The LiDAR super-resolution networks reconstruct the upscaled range image having test resolution. We measure the mean absolute error (MAE) of all the pixels in the predicted 2-D range images.  ... 
arXiv:2203.06413v1 fatcat:ho4gdae7fvazta7s37ncvu324a

Self-attention Negative Feedback Network for Real-time Image Super-Resolution

Xiangbin Liu, Shuqi Chen, Liping Song, Marcin Woźniak, Shuai Liu
2021 Journal of King Saud University: Computer and Information Sciences  
Therefore, this paper proposes a self-attention negative feedback network (SRAFBN) for realizing the real-time image SR.  ...  In the field of real-time image enhancement, image super-resolution (SR) is an important research hotspot.  ...  Super-resolution network flow chart of self-attention negative feedback image For instance, Luo et al.  ... 
doi:10.1016/j.jksuci.2021.07.014 fatcat:v733gwxrrzdldktupldau7jpge

PIRM2018 Challenge on Spectral Image Super-Resolution: Methods and Results [chapter]

Mehrdad Shoeiby, Antonio Robles-Kelly, Radu Timofte, Ruofan Zhou, Fayez Lahoud, Sabine Süsstrunk, Zhiwei Xiong, Zhan Shi, Chang Chen, Dong Liu, Zheng-Jun Zha, Feng Wu (+9 others)
2019 Lecture Notes in Computer Science  
The first of these (Track 1) is about example-based single spectral image super-resolution. The second one (Track 2) is on colour-guided spectral image super-resolution.  ...  In this manner, Track 1 focuses on the problem of super-resolving the spatial resolution of spectral images given training pairs of low and high spatial resolution spectral images.  ...  Introduction Image super-resolution (SR) aims at reconstructing details, that is high frequency information that was lost in an image due to various reasons such as camera sensor limitations, blurring,  ... 
doi:10.1007/978-3-030-11021-5_22 fatcat:2l64hl3bmffw5b7vyb26tv624m
« Previous Showing results 1 — 15 out of 26,193 results