PyTorchKR Papers

최신 논문

총 3,696편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2104.01530공식 코드 · Papers with Code

High-resolution Depth Maps Imaging via Attention-based Hierarchical Multi-modal Fusion

Depth map records distance between the viewpoint and objects in the scene, which plays a critical role in many real-world applications. However, depth map captured by consumer-grade RGB-D cameras suffers from low spatial resolution. Guided depth map super-resolution (DSR) is a popular approach to address this problem, which attempts to restore a high-resolution (HR) depth map from the input low-resolution (LR) depth and its coupled HR RGB image that serves as the guidance. The most challenging problems for guided DSR are how to correctly select consistent structures and propagate them, and properly handle inconsistent ones. In this paper, we propose a novel attention-based hierarchical multi-modal fusion (AHMF) network for guided DSR. Specifically, to effectively extract and combine relevant information from LR depth and HR guidance, we propose a multi-modal attention based fusion (MMAF) strategy for hierarchical convolutional layers, including a feature enhance block to select valuable features and a feature recalibration block to unify the similarity metrics of modalities with different appearance characteristics. Furthermore, we propose a bi-directional hierarchical feature collaboration (BHFC) module to fully leverage low-level spatial information and high-level structure information among multi-scale features. Experimental results show that our approach outperforms state-of-the-art methods in terms of reconstruction accuracy, running speed and memory efficiency.

arXiv:2104.00848공식 코드 · Papers with Code

SDAN: Squared Deformable Alignment Network for Learning Misaligned Optical Zoom

Deep Neural Network (DNN) based super-resolution algorithms have greatly improved the quality of the generated images. However, these algorithms often yield significant artifacts when dealing with real-world super-resolution problems due to the difficulty in learning misaligned optical zoom. In this paper, we introduce a Squared Deformable Alignment Network (SDAN) to address this issue. Our network learns squared per-point offsets for convolutional kernels, and then aligns features in corrected convolutional windows based on the offsets. So the misalignment will be minimized by the extracted aligned features. Different from the per-point offsets used in the vanilla Deformable Convolutional Network (DCN), our proposed squared offsets not only accelerate the offset learning but also improve the generation quality with fewer parameters. Besides, we further propose an efficient cross packing attention layer to boost the accuracy of the learned offsets. It leverages the packing and unpacking operations to enlarge the receptive field of the offset learning and to enhance the ability of extracting the spatial connection between the low-resolution images and the referenced images. Comprehensive experiments show the superiority of our method over other state-of-the-art methods in both computational efficiency and realistic details.

arXiv:2104.00100공식 코드 · Papers with Code

MR Slice Profile Estimation by Learning to Match Internal Patch Distributions

To super-resolve the through-plane direction of a multi-slice 2D magnetic resonance (MR) image, its slice selection profile can be used as the degeneration model from high resolution (HR) to low resolution (LR) to create paired data when training a supervised algorithm. Existing super-resolution algorithms make assumptions about the slice selection profile since it is not readily known for a given image. In this work, we estimate a slice selection profile given a specific image by learning to match its internal patch distributions. Specifically, we assume that after applying the correct slice selection profile, the image patch distribution along HR in-plane directions should match the distribution along the LR through-plane direction. Therefore, we incorporate the estimation of a slice selection profile as part of learning a generator in a generative adversarial network (GAN). In this way, the slice selection profile can be learned without any external data. Our algorithm was tested using simulations from isotropic MR images, incorporated in a through-plane super-resolution algorithm to demonstrate its benefits, and also used as a tool to measure image resolution. Our code is at https://github.com/shuohan/espreso2.

arXiv:2104.00024공식 코드 · Papers with Code

RetrievalFuse: Neural 3D Scene Reconstruction with a Database

3D reconstruction of large scenes is a challenging problem due to the high-complexity nature of the solution space, in particular for generative neural networks. In contrast to traditional generative learned models which encode the full generative process into a neural network and can struggle with maintaining local details at the scene level, we introduce a new method that directly leverages scene geometry from the training database. First, we learn to synthesize an initial estimate for a 3D scene, constructed by retrieving a top-k set of volumetric chunks from the scene database. These candidates are then refined to a final scene generation with an attention-based refinement that can effectively select the most consistent set of geometry from the candidates and combine them together to create an output scene, facilitating transfer of coherent structures and local detail from train scene geometry. We demonstrate our neural scene reconstruction with a database for the tasks of 3D super resolution and surface reconstruction from sparse point clouds, showing that our approach enables generation of more coherent, accurate 3D scenes, improving on average by over 8% in IoU over state-of-the-art scene reconstruction.

arXiv:2104.00416공식 코드 · Papers with Code

Unsupervised Degradation Representation Learning for Blind Super-Resolution

Most existing CNN-based super-resolution (SR) methods are developed based on an assumption that the degradation is fixed and known (e.g., bicubic downsampling). However, these methods suffer a severe performance drop when the real degradation is different from their assumption. To handle various unknown degradations in real-world applications, previous methods rely on degradation estimation to reconstruct the SR image. Nevertheless, degradation estimation methods are usually time-consuming and may lead to SR failure due to large estimation errors. In this paper, we propose an unsupervised degradation representation learning scheme for blind SR without explicit degradation estimation. Specifically, we learn abstract representations to distinguish various degradations in the representation space rather than explicit estimation in the pixel space. Moreover, we introduce a Degradation-Aware SR (DASR) network with flexible adaption to various degradations based on the learned representations. It is demonstrated that our degradation representation learning scheme can extract discriminative representations to obtain accurate degradation information. Experiments on both synthetic and real images show that our network achieves state-of-the-art performance for the blind SR task. Code is available at: https://github.com/LongguangWang/DASR.

arXiv:2103.17261

Video-Specific Autoencoders for Exploring, Editing and Transmitting Videos

We study video-specific autoencoders that allow a human user to explore, edit, and efficiently transmit videos. Prior work has independently looked at these problems (and sub-problems) and proposed different formulations. In this work, we train a simple autoencoder (from scratch) on multiple frames of a specific video. We observe: (1) latent codes learned by a video-specific autoencoder capture spatial and temporal properties of that video; and (2) autoencoders can project out-of-sample inputs onto the video-specific manifold. These two properties allow us to explore, edit, and efficiently transmit a video using one learned representation. For e.g., linear operations on latent codes allow users to visualize the contents of a video. Associating latent codes of a video and manifold projection enables users to make desired edits. Interpolating latent codes and manifold projection allows the transmission of sparse low-res frames over a network.

arXiv:2103.16935공식 코드 · Papers with Code

Near field Acoustic Holography on arbitrary shapes using Convolutional Neural Network

Near-field Acoustic Holography (NAH) is a well-known problem aimed at estimating the vibrational velocity field of a structure by means of acoustic measurements. In this paper, we propose a NAH technique based on Convolutional Neural Network (CNN). The devised CNN predicts the vibrational field on the surface of arbitrary shaped plates (violin plates) with orthotropic material properties from a limited number of measurements. In particular, the architecture, named Super Resolution CNN (SRCNN), is able to estimate the vibrational field with a higher spatial resolution compared to the input pressure. The pressure and velocity datasets have been generated through Finite Element Method simulations. We validate the proposed method by comparing the estimates with the synthesized ground truth and with a state-of-the-art technique. Moreover, we evaluate the robustness of the devised network against noisy input data.

arXiv:2103.15875

In-Place Scene Labelling and Understanding with Implicit Scene Representation

Semantic labelling is highly correlated with geometry and radiance reconstruction, as scene entities with similar shape and appearance are more likely to come from similar classes. Recent implicit neural reconstruction techniques are appealing as they do not require prior training data, but the same fully self-supervised approach is not possible for semantics because labels are human-defined properties. We extend neural radiance fields (NeRF) to jointly encode semantics with appearance and geometry, so that complete and accurate 2D semantic labels can be achieved using a small amount of in-place annotations specific to the scene. The intrinsic multi-view consistency and smoothness of NeRF benefit semantics by enabling sparse labels to efficiently propagate. We show the benefit of this approach when labels are either sparse or very noisy in room-scale scenes. We demonstrate its advantageous properties in various interesting applications such as an efficient scene labelling tool, novel semantic view synthesis, label denoising, super-resolution, label interpolation and multi-view semantic label fusion in visual semantic mapping systems.

arXiv:2103.15977공식 코드 · Papers with Code

Flow-based Kernel Prior with Application to Blind Super-Resolution

Kernel estimation is generally one of the key problems for blind image super-resolution (SR). Recently, Double-DIP proposes to model the kernel via a network architecture prior, while KernelGAN employs the deep linear network and several regularization losses to constrain the kernel space. However, they fail to fully exploit the general SR kernel assumption that anisotropic Gaussian kernels are sufficient for image SR. To address this issue, this paper proposes a normalizing flow-based kernel prior (FKP) for kernel modeling. By learning an invertible mapping between the anisotropic Gaussian kernel distribution and a tractable latent distribution, FKP can be easily used to replace the kernel modeling modules of Double-DIP and KernelGAN. Specifically, FKP optimizes the kernel in the latent space rather than the network parameter space, which allows it to generate reasonable kernel initialization, traverse the learned kernel manifold and improve the optimization stability. Extensive experiments on synthetic and real-world images demonstrate that the proposed FKP can significantly improve the kernel estimation accuracy with less parameters, runtime and memory usage, leading to state-of-the-art blind SR results.

arXiv:2012.08943

CT Super Resolution via Zero Shot Learning

Computed Tomography (CT) is an advanced imaging technology used in many important applications. Here we present a deep-learning (DL) based CT super-resolution (SR) method that can reconstruct low-resolution (LR) sinograms into high resolution (HR) CT images. The method synergistically combines a SR model in sinogram domain, a deblur model in image domain, and the iterative framework into a CT SR algorithm super resolution and deblur based iterative reconstruction (SADIR). We incorporated the CT domain knowledge into the SADIR and unrolled it into a DL network (SADIR Net). The SADIR Net is a zero shot learning (ZSL) network, which can be trained and tested with a single sinogram in the test time. The SADIR was evaluated via SR CT imaging of a Catphan700 physical phantom and a biological ham, and its performance was compared to the other state of the art (SotA) DL-based methods. The results show that the zero-shot SADIR-Net can indeed provide a performance comparable to the other SotA methods for CT SR reconstruction, especially in situations where training data is limited. The SADIR method can find use in improving CT resolution beyond hardware limits or lowering requirement on CT hardware.

arXiv:2103.11649

Gauging diffraction patterns: field of view and bandwidth estimation in lensless holography

The purpose of this work is to provide theoretically grounded assessment on both the field-of-view and the bandwidth of a lensless holographic setup. Indeed, while previous works have presented results with super-resolution and field-of-view extrapolation, there is no well established rules to determine them. We show that the theoretical field of view can be hugely large with a spatial-frequency bandwidth only limited by the wavelength leading to an unthinkable number of degrees of freedom. To keep a realistic field of view and bandwidth, we propose several practical bounds based on few setup properties: namely the noise level and the spatio-temporal coherence of the source.

arXiv:2103.11632

Mathematical Theory of Computational Resolution Limit in Multi-dimensions

Resolving a linear combination of point sources from their band-limited Fourier data is a fundamental problem in imaging and signal processing. With the incomplete Fourier data and the inevitable noise in the measurement, there is a fundamental limit on the separation distance between point sources that can be resolved. This is the so-called resolution limit problem. Characterization of this resolution limit is still a long-standing puzzle despite the prevalent use of the classic Rayleigh limit. It is well-known that Rayleigh limit is heuristic and its drawbacks become prominent when dealing with data that is subjected to delicate processing, as is what modern computational imaging methods do. Therefore, more precise characterization of the resolution limit becomes increasingly necessary with the development of data processing methods. For this purpose, we developed a theory of "computational resolution limit" for both number detection and support recovery in one dimension in [arXiv:2003.02917[cs.IT], arXiv:1912.05430[eess.IV]]. In this paper, we extend the one-dimensional theory to multi-dimensions. More precisely, we define and quantitatively characterize the "computational resolution limit" for the number detection and support recovery problems in a general k-dimensional space. Our results indicate that there exists a phase transition phenomenon regarding to the super-resolution factor and the signal-to-noise ratio in each of the two recovery problems. Our main results are derived using a subspace projection strategy. Finally, to verify the theory, we proposed deterministic subspace projection based algorithms for the number detection and support recovery problems in dimension two and three. The numerical results confirm the phase transition phenomenon predicted by the theory.

arXiv:2103.15295공식 코드 · Papers with Code

Best-Buddy GANs for Highly Detailed Image Super-Resolution

We consider the single image super-resolution (SISR) problem, where a high-resolution (HR) image is generated based on a low-resolution (LR) input. Recently, generative adversarial networks (GANs) become popular to hallucinate details. Most methods along this line rely on a predefined single-LR-single-HR mapping, which is not flexible enough for the SISR task. Also, GAN-generated fake details may often undermine the realism of the whole image. We address these issues by proposing best-buddy GANs (Beby-GAN) for rich-detail SISR. Relaxing the immutable one-to-one constraint, we allow the estimated patches to dynamically seek the best supervision during training, which is beneficial to producing more reasonable details. Besides, we propose a region-aware adversarial learning strategy that directs our model to focus on generating details for textured areas adaptively. Extensive experiments justify the effectiveness of our method. An ultra-high-resolution 4K dataset is also constructed to facilitate future super-resolution research.

arXiv:2103.15290공식 코드 · Papers with Code

Transitional Learning: Exploring the Transition States of Degradation for Blind Super-resolution

Being extremely dependent on iterative estimation of the degradation prior or optimization of the model from scratch, the existing blind super-resolution (SR) methods are generally time-consuming and less effective, as the estimation of degradation proceeds from a blind initialization and lacks interpretable degradation priors. To address it, this paper proposes a transitional learning method for blind SR using an end-to-end network without any additional iterations in inference, and explores an effective representation for unknown degradation. To begin with, we analyze and demonstrate the transitionality of degradations as interpretable prior information to indirectly infer the unknown degradation model, including the widely used additive and convolutive degradations. We then propose a novel Transitional Learning method for blind Super-Resolution (TLSR), by adaptively inferring a transitional transformation function to solve the unknown degradations without any iterative operations in inference. Specifically, the end-to-end TLSR network consists of a degree of transitionality (DoT) estimation network, a homogeneous feature extraction network, and a transitional learning module. Quantitative and qualitative evaluations on blind SR tasks demonstrate that the proposed TLSR achieves superior performances and costs fewer complexities against the state-of-the-art blind SR methods. The code is available at github.com/YuanfeiHuang/TLSR.

arXiv:2103.15683

Omniscient Video Super-Resolution

Most recent video super-resolution (SR) methods either adopt an iterative manner to deal with low-resolution (LR) frames from a temporally sliding window, or leverage the previously estimated SR output to help reconstruct the current frame recurrently. A few studies try to combine these two structures to form a hybrid framework but have failed to give full play to it. In this paper, we propose an omniscient framework to not only utilize the preceding SR output, but also leverage the SR outputs from the present and future. The omniscient framework is more generic because the iterative, recurrent and hybrid frameworks can be regarded as its special cases. The proposed omniscient framework enables a generator to behave better than its counterparts under other frameworks. Abundant experiments on public datasets show that our method is superior to the state-of-the-art methods in objective metrics, subjective visual effects and complexity. Our code will be made public.

arXiv:2103.14373공식 코드 · Papers with Code

D2C-SR: A Divergence to Convergence Approach for Real-World Image Super-Resolution

In this paper, we present D2C-SR, a novel framework for the task of real-world image super-resolution. As an ill-posed problem, the key challenge in super-resolution related tasks is there can be multiple predictions for a given low-resolution input. Most classical deep learning based approaches ignored the fundamental fact and lack explicit modeling of the underlying high-frequency distribution which leads to blurred results. Recently, some methods of GAN-based or learning super-resolution space can generate simulated textures but do not promise the accuracy of the textures which have low quantitative performance. Rethinking both, we learn the distribution of underlying high-frequency details in a discrete form and propose a two-stage pipeline: divergence stage to convergence stage. At divergence stage, we propose a tree-based structure deep network as our divergence backbone. Divergence loss is proposed to encourage the generated results from the tree-based network to diverge into possible high-frequency representations, which is our way of discretely modeling the underlying high-frequency distribution. At convergence stage, we assign spatial weights to fuse these divergent predictions to obtain the final output with more accurate details. Our approach provides a convenient end-to-end manner to inference. We conduct evaluations on several real-world benchmarks, including a new proposed D2CRealSR dataset with x8 scaling factor. Our experiments demonstrate that D2C-SR achieves better accuracy and visual improvements against state-of-the-art methods, with a significantly less parameters number and our D2C structure can also be applied as a generalized structure to some other methods to obtain improvement. Our codes and dataset are available at https://github.com/megvii-research/D2C-SR

arXiv:2103.14616

Training a Task-Specific Image Reconstruction Loss

The choice of a loss function is an important factor when training neural networks for image restoration problems, such as single image super resolution. The loss function should encourage natural and perceptually pleasing results. A popular choice for a loss is a pre-trained network, such as VGG, which is used as a feature extractor for computing the difference between restored and reference images. However, such an approach has multiple drawbacks: it is computationally expensive, requires regularization and hyper-parameter tuning, and involves a large network trained on an unrelated task. Furthermore, it has been observed that there is no single loss function that works best across all applications and across different datasets. In this work, we instead propose to train a set of loss functions that are application specific in nature. Our loss function comprises a series of discriminators that are trained to detect and penalize the presence of application-specific artifacts. We show that a single natural image and corresponding distortions are sufficient to train our feature extractor that outperforms state-of-the-art loss functions in applications like single image super resolution, denoising, and JPEG artifact removal. Finally, we conclude that an effective loss function does not have to be a good predictor of perceived image quality, but instead needs to be specialized in identifying the distortions for a given restoration method.

arXiv:2103.14006공식 코드 · Papers with Code

Designing a Practical Degradation Model for Deep Blind Image Super-Resolution

It is widely acknowledged that single image super-resolution (SISR) methods would not perform well if the assumed degradation model deviates from those in real images. Although several degradation models take additional factors into consideration, such as blur, they are still not effective enough to cover the diverse degradations of real images. To address this issue, this paper proposes to design a more complex but practical degradation model that consists of randomly shuffled blur, downsampling and noise degradations. Specifically, the blur is approximated by two convolutions with isotropic and anisotropic Gaussian kernels; the downsampling is randomly chosen from nearest, bilinear and bicubic interpolations; the noise is synthesized by adding Gaussian noise with different noise levels, adopting JPEG compression with different quality factors, and generating processed camera sensor noise via reverse-forward camera image signal processing (ISP) pipeline model and RAW image noise model. To verify the effectiveness of the new degradation model, we have trained a deep blind ESRGAN super-resolver and then applied it to super-resolve both synthetic and real images with diverse degradations. The experimental results demonstrate that the new degradation model can help to significantly improve the practicability of deep super-resolvers, thus providing a powerful alternative solution for real SISR applications.

arXiv:2103.13778

Multi-frame Super-resolution from Noisy Data

Obtaining high resolution images from low resolution data with clipped noise is algorithmically challenging due to the ill-posed nature of the problem. So far such problems have hardly been tackled, and the few existing approaches use simplistic regularisers. We show the usefulness of two adaptive regularisers based on anisotropic diffusion ideas: Apart from evaluating the classical edge-enhancing anisotropic diffusion regulariser, we introduce a novel non-local one with one-sided differences and superior performance. It is termed sector diffusion. We combine it with all six variants of the classical super-resolution observational model that arise from permutations of its three operators for warping, blurring, and downsampling. Surprisingly, the evaluation in a practically relevant noisy scenario produces a different ranking than the one in the noise-free setting in our previous work (SSVM 2017).

arXiv:2103.13676

JDSR-GAN: Constructing An Efficient Joint Learning Network for Masked Face Super-Resolution

With the growing importance of preventing the COVID-19 virus, face images obtained in most video surveillance scenarios are low resolution with mask simultaneously. However, most of the previous face super-resolution solutions can not handle both tasks in one model. In this work, we treat the mask occlusion as image noise and construct a joint and collaborative learning network, called JDSR-GAN, for the masked face super-resolution task. Given a low-quality face image with the mask as input, the role of the generator composed of a denoising module and super-resolution module is to acquire a high-quality high-resolution face image. The discriminator utilizes some carefully designed loss functions to ensure the quality of the recovered face images. Moreover, we incorporate the identity information and attention mechanism into our network for feasible correlated feature expression and informative feature learning. By jointly performing denoising and face super-resolution, the two tasks can complement each other and attain promising performance. Extensive qualitative and quantitative results show the superiority of our proposed JDSR-GAN over some comparable methods which perform the previous two tasks separately.