PyTorchKR Papers

최신 논문

총 3,696편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2112.10541

Implicit Neural Representation Learning for Hyperspectral Image Super-Resolution

Hyperspectral image (HSI) super-resolution without additional auxiliary image remains a constant challenge due to its high-dimensional spectral patterns, where learning an effective spatial and spectral representation is a fundamental issue. Recently, Implicit Neural Representations (INRs) are making strides as a novel and effective representation, especially in the reconstruction task. Therefore, in this work, we propose a novel HSI reconstruction model based on INR which represents HSI by a continuous function mapping a spatial coordinate to its corresponding spectral radiance values. In particular, as a specific implementation of INR, the parameters of parametric model are predicted by a hypernetwork that operates on feature extraction using convolution network. It makes the continuous functions map the spatial coordinates to pixel values in a content-aware manner. Moreover, periodic spatial encoding are deeply integrated with the reconstruction procedure, which makes our model capable of recovering more high frequency details. To verify the efficacy of our model, we conduct experiments on three HSI datasets (CAVE, NUS, and NTIRE2018). Experimental results show that the proposed model can achieve competitive reconstruction performance in comparison with the state-of-the-art methods. In addition, we provide an ablation study on the effect of individual components of our model. We hope this paper could server as a potent reference for future research.

arXiv:2112.09532공식 코드 · Papers with Code

Pixel Distillation: A New Knowledge Distillation Scheme for Low-Resolution Image Recognition

Previous knowledge distillation (KD) methods mostly focus on compressing network architectures, which is not thorough enough in deployment as some costs like transmission bandwidth and imaging equipment are related to the image size. Therefore, we propose Pixel Distillation that extends knowledge distillation into the input level while simultaneously breaking architecture constraints. Such a scheme can achieve flexible cost control for deployment, as it allows the system to adjust both network architecture and image quality according to the overall requirement of resources. Specifically, we first propose an input spatial representation distillation (ISRD) mechanism to transfer spatial knowledge from large images to student's input module, which can facilitate stable knowledge transfer between CNN and ViT. Then, a Teacher-Assistant-Student (TAS) framework is further established to disentangle pixel distillation into the model compression stage and input compression stage, which significantly reduces the overall complexity of pixel distillation and the difficulty of distilling intermediate knowledge. Finally, we adapt pixel distillation to object detection via an aligned feature for preservation (AFP) strategy for TAS, which aligns output dimensions of detectors at each stage by manipulating features and anchors of the assistant. Comprehensive experiments on image classification and object detection demonstrate the effectiveness of our method. Code is available at https://github.com/gyguo/PixelDistillation.

arXiv:2112.10046공식 코드 · Papers with Code

A-ESRGAN: Training Real-World Blind Super-Resolution with Attention U-Net Discriminators

Blind image super-resolution(SR) is a long-standing task in CV that aims to restore low-resolution images suffering from unknown and complex distortions. Recent work has largely focused on adopting more complicated degradation models to emulate real-world degradations. The resulting models have made breakthroughs in perceptual loss and yield perceptually convincing results. However, the limitation brought by current generative adversarial network structures is still significant: treating pixels equally leads to the ignorance of the image's structural features, and results in performance drawbacks such as twisted lines and background over-sharpening or blurring. In this paper, we present A-ESRGAN, a GAN model for blind SR tasks featuring an attention U-Net based, multi-scale discriminator that can be seamlessly integrated with other generators. To our knowledge, this is the first work to introduce attention U-Net structure as the discriminator of GAN to solve blind SR problems. And the paper also gives an interpretation for the mechanism behind multi-scale attention U-Net that brings performance breakthrough to the model. Through comparison experiments with prior works, our model presents state-of-the-art level performance on the non-reference natural image quality evaluator metric. And our ablation studies have shown that with our discriminator, the RRDB based generator can leverage the structural features of an image in multiple scales, and consequently yields more perceptually realistic high-resolution images compared to prior works.

arXiv:2112.08644

A comparative study of paired versus unpaired deep learning methods for physically enhancing digital rock image resolution

X-ray micro-computed tomography (micro-CT) has been widely leveraged to characterise pore-scale geometry in subsurface porous rock. Recent developments in super resolution (SR) methods using deep learning allow the digital enhancement of low resolution (LR) images over large spatial scales, creating SR images comparable to the high resolution (HR) ground truth. This circumvents traditional resolution and field-of-view trade-offs. An outstanding issue is the use of paired (registered) LR and HR data, which is often required in the training step of such methods but is difficult to obtain. In this work, we rigorously compare two different state-of-the-art SR deep learning techniques, using both paired and unpaired data, with like-for-like ground truth data. The first approach requires paired images to train a convolutional neural network (CNN) while the second approach uses unpaired images to train a generative adversarial network (GAN). The two approaches are compared using a micro-CT carbonate rock sample with complicated micro-porous textures. We implemented various image based and numerical verifications and experimental validation to quantitatively evaluate the physical accuracy and sensitivities of the two methods. Our quantitative results show that unpaired GAN approach can reconstruct super-resolution images as precise as paired CNN method, with comparable training times and dataset requirement. This unlocks new applications for micro-CT image enhancement using unpaired deep learning methods; image registration is no longer needed during the data processing stage. Decoupled images from data storage platforms can be exploited more efficiently to train networks for SR digital rock applications. This opens up a new pathway for various applications of multi-scale flow simulation in heterogeneous porous media.

arXiv:2112.08950공식 코드 · Papers with Code

Stable Long-Term Recurrent Video Super-Resolution

Recurrent models have gained popularity in deep learning (DL) based video super-resolution (VSR), due to their increased computational efficiency, temporal receptive field and temporal consistency compared to sliding-window based models. However, when inferring on long video sequences presenting low motion (i.e. in which some parts of the scene barely move), recurrent models diverge through recurrent processing, generating high frequency artifacts. To the best of our knowledge, no study about VSR pointed out this instability problem, which can be critical for some real-world applications. Video surveillance is a typical example where such artifacts would occur, as both the camera and the scene stay static for a long time. In this work, we expose instabilities of existing recurrent VSR networks on long sequences with low motion. We demonstrate it on a new long sequence dataset Quasi-Static Video Set, that we have created. Finally, we introduce a new framework of recurrent VSR networks that is both stable and competitive, based on Lipschitz stability theory. We propose a new recurrent VSR network, coined Middle Recurrent Video Super-Resolution (MRVSR), based on this framework. We empirically show its competitive performance on long sequences with low motion.

arXiv:2112.08655공식 코드 · Papers with Code

Feature Distillation Interaction Weighting Network for Lightweight Image Super-Resolution

Convolutional neural networks based single-image super-resolution (SISR) has made great progress in recent years. However, it is difficult to apply these methods to real-world scenarios due to the computational and memory cost. Meanwhile, how to take full advantage of the intermediate features under the constraints of limited parameters and calculations is also a huge challenge. To alleviate these issues, we propose a lightweight yet efficient Feature Distillation Interaction Weighted Network (FDIWN). Specifically, FDIWN utilizes a series of specially designed Feature Shuffle Weighted Groups (FSWG) as the backbone, and several novel mutual Wide-residual Distillation Interaction Blocks (WDIB) form an FSWG. In addition, Wide Identical Residual Weighting (WIRW) units and Wide Convolutional Residual Weighting (WCRW) units are introduced into WDIB for better feature distillation. Moreover, a Wide-Residual Distillation Connection (WRDC) framework and a Self-Calibration Fusion (SCF) unit are proposed to interact features with different scales more flexibly and efficiently.Extensive experiments show that our FDIWN is superior to other models to strike a good balance between model performance and efficiency. The code is available at https://github.com/IVIPLab/FDIWN.

arXiv:2112.07589

Mitigating Channel-wise Noise for Single Image Super Resolution

In practice, images can contain different amounts of noise for different color channels, which is not acknowledged by existing super-resolution approaches. In this paper, we propose to super-resolve noisy color images by considering the color channels jointly. Noise statistics are blindly estimated from the input low-resolution image and are used to assign different weights to different color channels in the data cost. Implicit low-rank structure of visual data is enforced via nuclear norm minimization in association with adaptive weights, which is added as a regularization term to the cost. Additionally, multi-scale details of the image are added to the model through another regularization term that involves projection onto PCA basis, which is constructed using similar patches extracted across different scales of the input image. The results demonstrate the super-resolving capability of the approach in real scenarios.

arXiv:2112.07315공식 코드 · Papers with Code

Kernel-aware Burst Blind Super-Resolution

Burst super-resolution (SR) technique provides a possibility of restoring rich details from low-quality images. However, since real world low-resolution (LR) images in practical applications have multiple complicated and unknown degradations, existing non-blind (e.g., bicubic) designed networks usually suffer severe performance drop in recovering high-resolution (HR) images. In this paper, we address the problem of reconstructing HR images from raw burst sequences acquired from a modern handheld device. The central idea is a kernel-guided strategy which can solve the burst SR problem with two steps: kernel estimation and HR image restoration. The former estimates burst kernels from raw inputs, while the latter predicts the super-resolved image based on the estimated kernels. Furthermore, we introduce a pyramid kernel-aware deformable alignment module which can effectively align the raw images with consideration of the blurry priors. Extensive experiments on synthetic and real-world datasets demonstrate that the proposed method can perform favorable state-of-the-art performance in the burst SR problem. Our codes are available at \url{https://github.com/shermanlian/KBNet}.

arXiv:2112.05756

Enhancing Multi-Scale Implicit Learning in Image Super-Resolution with Integrated Positional Encoding

Is the center position fully capable of representing a pixel? There is nothing wrong to represent pixels with their centers in a discrete image representation, but it makes more sense to consider each pixel as the aggregation of signals from a local area in an image super-resolution (SR) context. Despite the great capability of coordinate-based implicit representation in the field of arbitrary-scale image SR, this area's nature of pixels is not fully considered. To this end, we propose integrated positional encoding (IPE), extending traditional positional encoding by aggregating frequency information over the pixel area. We apply IPE to the state-of-the-art arbitrary-scale image super-resolution method: local implicit image function (LIIF), presenting IPE-LIIF. We show the effectiveness of IPE-LIIF by quantitative and qualitative evaluations, and further demonstrate the generalization ability of IPE to larger image scales and multiple implicit-based methods. Code will be released.

arXiv:2112.05755

Information Prebuilt Recurrent Reconstruction Network for Video Super-Resolution

The video super-resolution (VSR) method based on the recurrent convolutional network has strong temporal modeling capability for video sequences. However, the temporal receptive field of different recurrent units in the unidirectional recurrent network is unbalanced. Earlier reconstruction frames receive less spatio-temporal information, resulting in fuzziness or artifacts. Although the bidirectional recurrent network can alleviate this problem, it requires more memory space and fails to perform many tasks with low latency requirements. To solve the above problems, we propose an end-to-end information prebuilt recurrent reconstruction network (IPRRN), consisting of an information prebuilt network (IPNet) and a recurrent reconstruction network (RRNet). By integrating sufficient information from the front of the video to build the hidden state needed for the initially recurrent unit to help restore the earlier frames, the information prebuilt network balances the input information difference at different time steps. In addition, we demonstrate an efficient recurrent reconstruction network, which outperforms the existing unidirectional recurrent schemes in all aspects. Many experiments have verified the effectiveness of the network we propose, which can effectively achieve better quantitative and qualitative evaluation performance compared to the existing state-of-the-art methods.

arXiv:2112.06536

SphereSR: 360° Image Super-Resolution with Arbitrary Projection via Continuous Spherical Image Representation

The 360{\deg}imaging has recently gained great attention; however, its angular resolution is relatively lower than that of a narrow field-of-view (FOV) perspective image as it is captured by using fisheye lenses with the same sensor size. Therefore, it is beneficial to super-resolve a 360{\deg}image. Some attempts have been made but mostly considered the equirectangular projection (ERP) as one of the way for 360{\deg}image representation despite of latitude-dependent distortions. In that case, as the output high-resolution(HR) image is always in the same ERP format as the low-resolution (LR) input, another information loss may occur when transforming the HR image to other projection types. In this paper, we propose SphereSR, a novel framework to generate a continuous spherical image representation from an LR 360{\deg}image, aiming at predicting the RGB values at given spherical coordinates for super-resolution with an arbitrary 360{\deg}image projection. Specifically, we first propose a feature extraction module that represents the spherical data based on icosahedron and efficiently extracts features on the spherical surface. We then propose a spherical local implicit image function (SLIIF) to predict RGB values at the spherical coordinates. As such, SphereSR flexibly reconstructs an HR image under an arbitrary projection type. Experiments on various benchmark datasets show that our method significantly surpasses existing methods.

arXiv:2112.06242공식 코드 · Papers with Code

Formulating Event-based Image Reconstruction as a Linear Inverse Problem with Deep Regularization using Optical Flow

Event cameras are novel bio-inspired sensors that measure per-pixel brightness differences asynchronously. Recovering brightness from events is appealing since the reconstructed images inherit the high dynamic range (HDR) and high-speed properties of events; hence they can be used in many robotic vision applications and to generate slow-motion HDR videos. However, state-of-the-art methods tackle this problem by training an event-to-image Recurrent Neural Network (RNN), which lacks explainability and is difficult to tune. In this work we show, for the first time, how tackling the combined problem of motion and brightness estimation leads us to formulate event-based image reconstruction as a linear inverse problem that can be solved without training an image reconstruction RNN. Instead, classical and learning-based regularizers are used to solve the problem and remove artifacts from the reconstructed images. The experiments show that the proposed approach generates images with visual quality on par with state-of-the-art methods despite only using data from a short time interval. State-of-the-art results are achieved using an image denoising Convolutional Neural Network (CNN) as the regularization function. The proposed regularized formulation and solvers have a unifying character because they can be applied also to reconstruct brightness from the second derivative. Additionally, the formulation is attractive because it can be naturally combined with super-resolution, motion-segmentation and color demosaicing. Code is available at https://github.com/tub-rip/event_based_image_rec_inverse_problem

arXiv:2112.06174공식 코드 · Papers with Code

Implicit Transformer Network for Screen Content Image Continuous Super-Resolution

Nowadays, there is an explosive growth of screen contents due to the wide application of screen sharing, remote cooperation, and online education. To match the limited terminal bandwidth, high-resolution (HR) screen contents may be downsampled and compressed. At the receiver side, the super-resolution (SR) of low-resolution (LR) screen content images (SCIs) is highly demanded by the HR display or by the users to zoom in for detail observation. However, image SR methods mostly designed for natural images do not generalize well for SCIs due to the very different image characteristics as well as the requirement of SCI browsing at arbitrary scales. To this end, we propose a novel Implicit Transformer Super-Resolution Network (ITSRN) for SCISR. For high-quality continuous SR at arbitrary ratios, pixel values at query coordinates are inferred from image features at key coordinates by the proposed implicit transformer and an implicit position encoding scheme is proposed to aggregate similar neighboring pixel values to the query one. We construct benchmark SCI1K and SCI1K-compression datasets with LR and HR SCI pairs. Extensive experiments show that the proposed ITSRN significantly outperforms several competitive continuous and discrete SR methods for both compressed and uncompressed SCIs.

arXiv:2112.05146

Come-Closer-Diffuse-Faster: Accelerating Conditional Diffusion Models for Inverse Problems through Stochastic Contraction

Diffusion models have recently attained significant interest within the community owing to their strong performance as generative models. Furthermore, its application to inverse problems have demonstrated state-of-the-art performance. Unfortunately, diffusion models have a critical downside - they are inherently slow to sample from, needing few thousand steps of iteration to generate images from pure Gaussian noise. In this work, we show that starting from Gaussian noise is unnecessary. Instead, starting from a single forward diffusion with better initialization significantly reduces the number of sampling steps in the reverse conditional diffusion. This phenomenon is formally explained by the contraction theory of the stochastic difference equations like our conditional diffusion strategy - the alternating applications of reverse diffusion followed by a non-expansive data consistency step. The new sampling strategy, dubbed Come-Closer-Diffuse-Faster (CCDF), also reveals a new insight on how the existing feed-forward neural network approaches for inverse problems can be synergistically combined with the diffusion models. Experimental results with super-resolution, image inpainting, and compressed sensing MRI demonstrate that our method can achieve state-of-the-art reconstruction performance at significantly reduced sampling steps.

arXiv:2112.04488공식 코드 · Papers with Code

A Dynamic Residual Self-Attention Network for Lightweight Single Image Super-Resolution

Deep learning methods have shown outstanding performance in many applications, including single-image super-resolution (SISR). With residual connection architecture, deeply stacked convolutional neural networks provide a substantial performance boost for SISR, but their huge parameters and computational loads are impractical for real-world applications. Thus, designing lightweight models with acceptable performance is one of the major tasks in current SISR research. The objective of lightweight network design is to balance a computational load and reconstruction performance. Most of the previous methods have manually designed complex and predefined fixed structures, which generally required a large number of experiments and lacked flexibility in the diversity of input image statistics. In this paper, we propose a dynamic residual self-attention network (DRSAN) for lightweight SISR, while focusing on the automated design of residual connections between building blocks. The proposed DRSAN has dynamic residual connections based on dynamic residual attention (DRA), which adaptively changes its structure according to input statistics. Specifically, we propose a dynamic residual module that explicitly models the DRA by finding the interrelation between residual paths and input image statistics, as well as assigning proper weights to each residual path. We also propose a residual self-attention (RSA) module to further boost the performance, which produces 3-dimensional attention maps without additional parameters by cooperating with residual structures. The proposed dynamic scheme, exploiting the combination of DRA and RSA, shows an efficient trade-off between computational complexity and network performance. Experimental results show that the DRSAN performs better than or comparable to existing state-of-the-art lightweight models for SISR.

arXiv:2112.04211

$\boldsymbolγ$-Net: Superresolving SAR Tomographic Inversion via Deep Learning

Synthetic aperture radar tomography (TomoSAR) has been extensively employed in 3-D reconstruction in dense urban areas using high-resolution SAR acquisitions. Compressive sensing (CS)-based algorithms are generally considered as the state of the art in super-resolving TomoSAR, in particular in the single look case. This superior performance comes at the cost of extra computational burdens, because of the sparse reconstruction, which cannot be solved analytically and we need to employ computationally expensive iterative solvers. In this paper, we propose a novel deep learning-based super-resolving TomoSAR inversion approach, $\boldsymbol{\gamma}$-Net, to tackle this challenge. $\boldsymbol{\gamma}$-Net adopts advanced complex-valued learned iterative shrinkage thresholding algorithm (CV-LISTA) to mimic the iterative optimization step in sparse reconstruction. Simulations show the height estimate from a well-trained $\boldsymbol{\gamma}$-Net approaches the Cram\'er-Rao lower bound while improving the computational efficiency by 1 to 2 orders of magnitude comparing to the first-order CS-based methods. It also shows no degradation in the super-resolution power comparing to the state-of-the-art second-order TomoSAR solvers, which are much more computationally expensive than the first-order methods. Specifically, $\boldsymbol{\gamma}$-Net reaches more than $90\%$ detection rate in moderate super-resolving cases at 25 measurements at 6dB SNR. Moreover, simulation at limited baselines demonstrates that the proposed algorithm outperforms the second-order CS-based method by a fair margin. Test on real TerraSAR-X data with just 6 interferograms also shows high-quality 3-D reconstruction with high-density detected double scatterers.

arXiv:2112.02270

Feature-based Recognition Framework for Super-resolution Images

In practical application, the performance of recognition network usually decreases when being applied on super-resolution images. In this paper, we propose a feature-based recognition network combined with GAN (FGAN). Our network improves the recognition accuracy by extracting more features that benefit recognition from SR images. In the experiment, we build three datasets using three different super-resolution algorithm, and our network increases the recognition accuracy by more than 6% comparing with ReaNet50 and DenseNet121.

arXiv:2112.02869공식 코드 · Papers with Code

Physics Driven Deep Retinex Fusion for Adaptive Infrared and Visible Image Fusion

Convolutional neural networks have turned into an illustrious tool for image fusion and super-resolution. However, their excellent performance cannot work without large fixed-paired datasets; and additionally, these high-demanded ground truth data always cannot be obtained easily in fusion tasks. In this study, we show that, the structures of generative networks capture a great deal of image feature priors, and then these priors are sufficient to reconstruct high-quality fused super-resolution result using only low-resolution inputs. By this way, we propose a novel self-supervised dataset-free method for adaptive infrared (IR) and visible (VIS) image super-resolution fusion named Deep Retinex Fusion (DRF). The key idea of DRF is first generating component priors which are disentangled from physical model using our designed generative networks ZipperNet, LightingNet and AdjustingNet, then combining these priors which captured by networks via adaptive fusion loss functions based on Retinex theory, and finally reconstructing the super-resolution fusion results. Furthermore, in order to verify the effectiveness of our reported DRF, both qualitative and quantitative experiments via comparing with other state-of-the-art methods are performed using different test sets. These results prove that, comparing with large datasets trained methods, DRF which works without any dataset achieves the best super-resolution fusion performance; and more importantly, DRF can adaptively balance IR and VIS information and has good noise immunity. DRF codes are open source available at https://github.com/GuYuanjie/Deep-Retinex-fusion.

arXiv:2112.02828공식 코드 · Papers with Code

PP-MSVSR: Multi-Stage Video Super-Resolution

Different from the Single Image Super-Resolution(SISR) task, the key for Video Super-Resolution(VSR) task is to make full use of complementary information across frames to reconstruct the high-resolution sequence. Since images from different frames with diverse motion and scene, accurately aligning multiple frames and effectively fusing different frames has always been the key research work of VSR tasks. To utilize rich complementary information of neighboring frames, in this paper, we propose a multi-stage VSR deep architecture, dubbed as PP-MSVSR, with local fusion module, auxiliary loss and re-align module to refine the enhanced result progressively. Specifically, in order to strengthen the fusion of features across frames in feature propagation, a local fusion module is designed in stage-1 to perform local feature fusion before feature propagation. Moreover, we introduce an auxiliary loss in stage-2 to make the features obtained by the propagation module reserve more correlated information connected to the HR space, and introduce a re-align module in stage-3 to make full use of the feature information of the previous stage. Extensive experiments substantiate that PP-MSVSR achieves a promising performance of Vid4 datasets, which achieves a PSNR of 28.13dB with only 1.45M parameters. And the PP-MSVSR-L exceeds all state of the art method on REDS4 datasets with considerable parameters. Code and models will be released in PaddleGAN\footnote{https://github.com/PaddlePaddle/PaddleGAN.}.

arXiv:2112.02719

A Survey on Deep learning based Document Image Enhancement

Digitized documents such as scientific articles, tax forms, invoices, contract papers, historic texts are widely used nowadays. These document images could be degraded or damaged due to various reasons including poor lighting conditions, shadow, distortions like noise and blur, aging, ink stain, bleed-through, watermark, stamp, etc. Document image enhancement plays a crucial role as a pre-processing step in many automated document analysis and recognition tasks such as character recognition. With recent advances in deep learning, many methods are proposed to enhance the quality of these document images. In this paper, we review deep learning-based methods, datasets, and metrics for six main document image enhancement tasks, including binarization, debluring, denoising, defading, watermark removal, and shadow removal. We summarize the recent works for each task and discuss their features, challenges, and limitations. We introduce multiple document image enhancement tasks that have received little to no attention, including over and under exposure correction, super resolution, and bleed-through removal. We identify several promising research directions and opportunities for future research.