PyTorchKR Papers

최신 논문

총 3,696편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2211.08428

CaDM: Codec-aware Diffusion Modeling for Neural-enhanced Video Streaming

Recent years have witnessed the dramatic growth of Internet video traffic, where the video bitstreams are often compressed and delivered in low quality to fit the streamer's uplink bandwidth. To alleviate the quality degradation, it comes the rise of Neural-enhanced Video Streaming (NVS), which shows great prospects for recovering low-quality videos by mostly deploying neural super-resolution (SR) on the media server. Despite its benefit, we reveal that current mainstream works with SR enhancement have not achieved the desired rate-distortion trade-off between bitrate saving and quality restoration, due to: (1) overemphasizing the enhancement on the decoder side while omitting the co-design of encoder, (2) limited generative capacity to recover high-fidelity perceptual details, and (3) optimizing the compression-and-restoration pipeline from the resolution perspective solely, without considering color bit-depth. Aiming at overcoming these limitations, we are the first to conduct an encoder-decoder (i.e., codec) synergy by leveraging the inherent visual-generative property of diffusion models. Specifically, we present the Codec-aware Diffusion Modeling (CaDM), a novel NVS paradigm to significantly reduce streaming delivery bitrates while holding pretty higher restoration capacity over existing methods. First, CaDM improves the encoder's compression efficiency by simultaneously reducing resolution and color bit-depth of video frames. Second, CaDM empowers the decoder with high-quality enhancement by making the denoising diffusion restoration aware of encoder's resolution-color conditions. Evaluation on public cloud services with OpenMMLab benchmarks shows that CaDM effectively saves up to 5.12 - 21.44 times bitrates based on common video standards and achieves much better recovery quality (e.g., FID of 0.61) over state-of-the-art neural-enhancing methods.

arXiv:2211.08703

SATVSR: Scenario Adaptive Transformer for Cross Scenarios Video Super-Resolution

Video Super-Resolution (VSR) aims to recover sequences of high-resolution (HR) frames from low-resolution (LR) frames. Previous methods mainly utilize temporally adjacent frames to assist the reconstruction of target frames. However, in the real world, there is a lot of irrelevant information in adjacent frames of videos with fast scene switching, these VSR methods cannot adaptively distinguish and select useful information. In contrast, with a transformer structure suitable for temporal tasks, we devise a novel adaptive scenario video super-resolution method. Specifically, we use optical flow to label the patches in each video frame, only calculate the attention of patches with the same label. Then select the most relevant label among them to supplement the spatial-temporal information of the target frame. This design can directly make the supplementary information come from the same scene as much as possible. We further propose a cross-scale feature aggregation module to better handle the scale variation problem. Compared with other video super-resolution methods, our method not only achieves significant performance gains on single-scene videos but also has better robustness on cross-scene datasets.

arXiv:2211.08658공식 코드 · Papers with Code

Consistent Direct Time-of-Flight Video Depth Super-Resolution

Direct time-of-flight (dToF) sensors are promising for next-generation on-device 3D sensing. However, limited by manufacturing capabilities in a compact module, the dToF data has a low spatial resolution (e.g., $\sim 20\times30$ for iPhone dToF), and it requires a super-resolution step before being passed to downstream tasks. In this paper, we solve this super-resolution problem by fusing the low-resolution dToF data with the corresponding high-resolution RGB guidance. Unlike the conventional RGB-guided depth enhancement approaches, which perform the fusion in a per-frame manner, we propose the first multi-frame fusion scheme to mitigate the spatial ambiguity resulting from the low-resolution dToF imaging. In addition, dToF sensors provide unique depth histogram information for each local patch, and we incorporate this dToF-specific feature in our network design to further alleviate spatial ambiguity. To evaluate our models on complex dynamic indoor environments and to provide a large-scale dToF sensor dataset, we introduce DyDToF, the first synthetic RGB-dToF video dataset that features dynamic objects and a realistic dToF simulator following the physical imaging process. We believe the methods and dataset are beneficial to a broad community as dToF depth sensing is becoming mainstream on mobile devices. Our code and data are publicly available: https://github.com/facebookresearch/DVSR/

arXiv:2211.07583

Nondestructive thermographic detection of internal defects using pixel-pattern based laser excitation and photothermal super resolution reconstruction

In this work, we present a novel approach to photothermal super resolution based thermographic resolution of internal defects using two-dimensional pixel pattern-based active photothermal laser heating in conjunction with subsequent numerical reconstruction to achieve a high-resolution reconstruction of internal defect structures. With the proposed adoption of pixelated patterns generated using laser coupled high-power DLP projector technology the complexity for achieving true two-dimensional super resolution can be dramatically reduced taking a crucial step forward towards widespread practical viability. Furthermore, based on the latest developments in high-power DLP projectors, we present their first application for structured pulsed thermographic inspection of macroscopic metal samples. In addition, a forward solution to the underlying inverse problem is proposed along with an appropriate heuristic to find the regularization parameters necessary for the numerical inversion in a laboratory setting. This allows the generation of synthetic measurement data, opening the door for the application of machine learning based methods for future improvements towards full automation of the method. Finally, the proposed method is experimentally validated and shown to outperform several established conventional thermographic testing techniques while conservatively improving the required measurement times by a factor of 8 compared to currently available photothermal super resolution techniques.

arXiv:2211.06958

Two-dimensional gridless super-resolution method for ISAR imaging

We are focused on improving the resolution of images of moving targets in Inverse Synthetic Aperture Radar (ISAR) imaging. This could be achieved by recovering the scattering points of a target that have stronger reflections than other target points, leading to increasing the higher Radar Cross Section (RCS) of a target. These points, however, are sparse and when the received data is incomplete, moving targets would not be properly recognizable in ISAR images. To increase the resolution in ISAR imaging, we propose the 2-Dimensional Reweighted Trace Minimization (2D-RWTM) method to retrieve frequencies of sparse scattering points in both range and cross-range directions. This method is a gridless super-resolution method, which does not depend on fitting the scattering point on the grids, leading to less complexity compared to the other methods. Using computer simulations, the proposed 2D-RWTM is compared to the Atomic Norm Minimization (ANM) in terms of the Mean Squared Errors (MSE). The results show that using the proposed method, the scattering points of a target are successfully recovered. It is shown that by selecting different weighting matrices and scattering points adjacent to each other, the recovery in ISAR imaging is still successful.

arXiv:2211.07555

Contrastive Learning for Climate Model Bias Correction and Super-Resolution

Climate models often require post-processing in order to make accurate estimates of local climate risk. The most common post-processing applied is bias-correction and spatial resolution enhancement. However, the statistical methods typically used for this not only are incapable of capturing multivariate spatial correlation information but are also reliant on rich observational data often not available outside of developed countries, limiting their potential. Here we propose an alternative approach to this challenge based on a combination of image super resolution (SR) and contrastive learning generative adversarial networks (GANs). We benchmark performance against NASA's flagship post-processed CMIP6 climate model product, NEX-GDDP. We find that our model successfully reaches a spatial resolution double that of NASA's product while also achieving comparable or improved levels of bias correction in both daily precipitation and temperature. The resulting higher fidelity simulations of present and forward-looking climate can enable more local, accurate models of hazards like flooding, drought, and heatwaves.

arXiv:2211.07302공식 코드 · Papers with Code

MedleyVox: An Evaluation Dataset for Multiple Singing Voices Separation

Separation of multiple singing voices into each voice is a rarely studied area in music source separation research. The absence of a benchmark dataset has hindered its progress. In this paper, we present an evaluation dataset and provide baseline studies for multiple singing voices separation. First, we introduce MedleyVox, an evaluation dataset for multiple singing voices separation. We specify the problem definition in this dataset by categorizing it into i) unison, ii) duet, iii) main vs. rest, and iv) N-singing separation. Second, to overcome the absence of existing multi-singing datasets for a training purpose, we present a strategy for construction of multiple singing mixtures using various single-singing datasets. Third, we propose the improved super-resolution network (iSRNet), which greatly enhances initial estimates of separation networks. Jointly trained with the Conv-TasNet and the multi-singing mixture construction strategy, the proposed iSRNet achieved comparable performance to ideal time-frequency masks on duet and unison subsets of MedleyVox. Audio samples, the dataset, and codes are available on our website (https://github.com/jeonchangbin49/MedleyVox).

arXiv:2211.07286공식 코드 · Papers with Code

CurvPnP: Plug-and-play Blind Image Restoration with Deep Curvature Denoiser

Due to the development of deep learning-based denoisers, the plug-and-play strategy has achieved great success in image restoration problems. However, existing plug-and-play image restoration methods are designed for non-blind Gaussian denoising such as zhang et al (2022), the performance of which visibly deteriorate for unknown noises. To push the limits of plug-and-play image restoration, we propose a novel framework with blind Gaussian prior, which can deal with more complicated image restoration problems in the real world. More specifically, we build up a new image restoration model by regarding the noise level as a variable, which is implemented by a two-stage blind Gaussian denoiser consisting of a noise estimation subnetwork and a denoising subnetwork, where the noise estimation subnetwork provides the noise level to the denoising subnetwork for blind noise removal. We also introduce the curvature map into the encoder-decoder architecture and the supervised attention module to achieve a highly flexible and effective convolutional neural network. The experimental results on image denoising, deblurring and single-image super-resolution are provided to demonstrate the advantages of our deep curvature denoiser and the resulting plug-and-play blind image restoration method over the state-of-the-art model-based and learning-based methods. Our model is shown to be able to recover the fine image details and tiny structures even when the noise level is unknown for different image restoration tasks. The source codes are available at https://github.com/Duanlab123/CurvPnP.

arXiv:2211.05910

Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 challenge: Report

Image super-resolution is a common task on mobile and IoT devices, where one often needs to upscale and enhance low-resolution images and video frames. While numerous solutions have been proposed for this problem in the past, they are usually not compatible with low-power mobile NPUs having many computational and memory constraints. In this Mobile AI challenge, we address this problem and propose the participants to design an efficient quantized image super-resolution solution that can demonstrate a real-time performance on mobile NPUs. The participants were provided with the DIV2K dataset and trained INT8 models to do a high-quality 3X image upscaling. The runtime of all models was evaluated on the Synaptics VS680 Smart Home board with a dedicated edge NPU capable of accelerating quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 60 FPS rate when reconstructing Full HD resolution images. A detailed description of all models developed in the challenge is provided in this paper.

arXiv:2211.06004

A Comprehensive Survey of Transformers for Computer Vision

As a special type of transformer, Vision Transformers (ViTs) are used to various computer vision applications (CV), such as image recognition. There are several potential problems with convolutional neural networks (CNNs) that can be solved with ViTs. For image coding tasks like compression, super-resolution, segmentation, and denoising, different variants of the ViTs are used. The purpose of this survey is to present the first application of ViTs in CV. The survey is the first of its kind on ViTs for CVs to the best of our knowledge. In the first step, we classify different CV applications where ViTs are applicable. CV applications include image classification, object detection, image segmentation, image compression, image super-resolution, image denoising, and anomaly detection. Our next step is to review the state-of-the-art in each category and list the available models. Following that, we present a detailed analysis and comparison of each model and list its pros and cons. After that, we present our insights and lessons learned for each category. Moreover, we discuss several open research challenges and future research directions.

arXiv:2211.05360

SRNR: Training neural networks for Super-Resolution MRI using Noisy high-resolution Reference data

Neural network (NN) based approaches for super-resolution MRI typically require high-SNR high-resolution reference data acquired in many subjects, which is time consuming and a barrier to feasible and accessible implementation. We propose to train NNs for Super-Resolution using Noisy Reference data (SRNR), leveraging the mechanism of the classic NN-based denoising method Noise2Noise. We systematically demonstrate that results from NNs trained using noisy and high-SNR references are similar for both simulated and empirical data. SRNR suggests a smaller number of repetitions of high-resolution reference data can be used to simplify the training data preparation for super-resolution MRI.

arXiv:2211.05256

Power Efficient Video Super-Resolution on Mobile NPUs with Deep Learning, Mobile AI & AIM 2022 challenge: Report

Video super-resolution is one of the most popular tasks on mobile devices, being widely used for an automatic improvement of low-bitrate and low-resolution video streams. While numerous solutions have been proposed for this problem, they are usually quite computationally demanding, demonstrating low FPS rates and power efficiency on mobile devices. In this Mobile AI challenge, we address this problem and propose the participants to design an end-to-end real-time video super-resolution solution for mobile NPUs optimized for low energy consumption. The participants were provided with the REDS training dataset containing video sequences for a 4X video upscaling task. The runtime and power efficiency of all models was evaluated on the powerful MediaTek Dimensity 9000 platform with a dedicated AI processing unit capable of accelerating floating-point and quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 500 FPS rate and 0.2 [Watt / 30 FPS] power consumption. A detailed description of all models developed in the challenge is provided in this paper.

arXiv:2211.05018공식 코드 · Papers with Code

The Best of Both Worlds: a Framework for Combining Degradation Prediction with High Performance Super-Resolution Networks

To date, the best-performing blind super-resolution (SR) techniques follow one of two paradigms: A) generate and train a standard SR network on synthetic low-resolution - high-resolution (LR - HR) pairs or B) attempt to predict the degradations an LR image has suffered and use these to inform a customised SR network. Despite significant progress, subscribers to the former miss out on useful degradation information that could be used to improve the SR process. On the other hand, followers of the latter rely on weaker SR networks, which are significantly outperformed by the latest architectural advancements. In this work, we present a framework for combining any blind SR prediction mechanism with any deep SR network, using a metadata insertion block to insert prediction vectors into SR network feature maps. Through comprehensive testing, we prove that state-of-the-art contrastive and iterative prediction schemes can be successfully combined with high-performance SR networks such as RCAN and HAN within our framework. We show that our hybrid models consistently achieve stronger SR performance than both their non-blind and blind counterparts. Furthermore, we demonstrate our framework's robustness by predicting degradations and super-resolving images from a complex pipeline of blurring, noise and compression.

arXiv:2211.04203

RRSR:Reciprocal Reference-based Image Super-Resolution with Progressive Feature Alignment and Selection

Reference-based image super-resolution (RefSR) is a promising SR branch and has shown great potential in overcoming the limitations of single image super-resolution. While previous state-of-the-art RefSR methods mainly focus on improving the efficacy and robustness of reference feature transfer, it is generally overlooked that a well reconstructed SR image should enable better SR reconstruction for its similar LR images when it is referred to as. Therefore, in this work, we propose a reciprocal learning framework that can appropriately leverage such a fact to reinforce the learning of a RefSR network. Besides, we deliberately design a progressive feature alignment and selection module for further improving the RefSR task. The newly proposed module aligns reference-input images at multi-scale feature spaces and performs reference-aware feature selection in a progressive manner, thus more precise reference features can be transferred into the input features and the network capability is enhanced. Our reciprocal learning paradigm is model-agnostic and it can be applied to arbitrary RefSR models. We empirically show that multiple recent state-of-the-art RefSR models can be consistently improved with our reciprocal learning paradigm. Furthermore, our proposed model together with the reciprocal learning strategy sets new state-of-the-art performances on multiple benchmarks.

arXiv:2211.02854

Rate-Distortion Optimized Post-Training Quantization for Learned Image Compression

Quantizing a floating-point neural network to its fixed-point representation is crucial for Learned Image Compression (LIC) because it improves decoding consistency for interoperability and reduces space-time complexity for implementation. Existing solutions often have to retrain the network for model quantization, which is time-consuming and impractical to some extent. This work suggests using Post-Training Quantization (PTQ) to process pretrained, off-the-shelf LIC models. We theoretically prove that minimizing quantization-induced mean square error (MSE) of model parameters (e.g., weight, bias, and activation) in PTQ is sub-optimal for compression tasks and thus develop a novel Rate-Distortion (R-D) Optimized PTQ (RDO-PTQ) to best retain the compression performance. Given a LIC model, RDO-PTQ layer-wisely determines the quantization parameters to transform the original floating-point parameters in 32-bit precision (FP32) to fixed-point ones at 8-bit precision (INT8), for which a tiny calibration image set is compressed in optimization to minimize R-D loss. Experiments reveal the outstanding efficiency of the proposed method on different LICs, showing the closest coding performance to their floating-point counterparts. Our method is a lightweight and plug-and-play approach without retraining model parameters but just adjusting quantization parameters, which is attractive to practitioners. Such an RDO-PTQ is a task-oriented PTQ scheme, which is then extended to quantize popular super-resolution and image classification models with negligible performance loss, further evidencing the generalization of our methodology. Related materials will be released at https://njuvision.github.io/RDO-PTQ.

arXiv:2211.03550

Underwater Image Super-Resolution using Generative Adversarial Network-based Model

Single image super-resolution (SISR) models are able to enhance the resolution and visual quality of underwater images and contribute to a better understanding of underwater environments. The integration of these models in Autonomous Underwater Vehicles (AUVs) can improve their performance in vision-based tasks. Real-Enhanced Super-Resolution Generative Adversarial Network (Real-ESRGAN) is an efficient model that has shown remarkable performance among SISR models. In this paper, we fine-tune the pre-trained Real-ESRGAN model for underwater image super-resolution. To fine-tune and evaluate the performance of the model, we use the USR-248 dataset. The fine-tuned model produces more realistic images with better visual quality compared to the Real-ESRGAN model.

arXiv:2211.03177

Measurement-Consistent Networks via a Deep Implicit Layer for Solving Inverse Problems

End-to-end deep neural networks (DNNs) have become the state-of-the-art (SOTA) for solving inverse problems. Despite their outstanding performance, during deployment, such networks are sensitive to minor variations in the testing pipeline and often fail to reconstruct small but important details, a feature critical in medical imaging, astronomy, or defence. Such instabilities in DNNs can be explained by the fact that they ignore the forward measurement model during deployment, and thus fail to enforce consistency between their output and the input measurements. To overcome this, we propose a framework that transforms any DNN for inverse problems into a measurement-consistent one. This is done by appending to it an implicit layer (or deep equilibrium network) designed to solve a model-based optimization problem. The implicit layer consists of a shallow learnable network that can be integrated into the end-to-end training while keeping the SOTA DNN fixed. Experiments on single-image super-resolution show that the proposed framework leads to significant improvements in reconstruction quality and robustness over the SOTA DNNs.

arXiv:2211.02973

Mixture-Net: Low-Rank Deep Image Prior Inspired by Mixture Models for Spectral Image Recovery

This paper proposes a non-data-driven deep neural network for spectral image recovery problems such as denoising, single hyperspectral image super-resolution, and compressive spectral imaging reconstruction. Unlike previous methods, the proposed approach, dubbed Mixture-Net, implicitly learns the prior information through the network. Mixture-Net consists of a deep generative model whose layers are inspired by the linear and non-linear low-rank mixture models, where the recovered image is composed of a weighted sum between the linear and non-linear decomposition. Mixture-Net also provides a low-rank decomposition interpreted as the spectral image abundances and endmembers, helpful in achieving remote sensing tasks without running additional routines. The experiments show the MixtureNet effectiveness outperforming state-of-the-art methods in recovery quality with the advantage of architecture interpretability.

arXiv:2211.02831공식 코드 · Papers with Code

A Survey of Deep Face Restoration: Denoise, Super-Resolution, Deblur, Artifact Removal

Face Restoration (FR) aims to restore High-Quality (HQ) faces from Low-Quality (LQ) input images, which is a domain-specific image restoration problem in the low-level computer vision area. The early face restoration methods mainly use statistic priors and degradation models, which are difficult to meet the requirements of real-world applications in practice. In recent years, face restoration has witnessed great progress after stepping into the deep learning era. However, there are few works to study deep learning-based face restoration methods systematically. Thus, this paper comprehensively surveys recent advances in deep learning techniques for face restoration. Specifically, we first summarize different problem formulations and analyze the characteristic of the face image. Second, we discuss the challenges of face restoration. Concerning these challenges, we present a comprehensive review of existing FR methods, including prior based methods and deep learning-based methods. Then, we explore developed techniques in the task of FR covering network architectures, loss functions, and benchmark datasets. We also conduct a systematic benchmark evaluation on representative methods. Finally, we discuss future directions, including network designs, metrics, benchmark datasets, applications,etc. We also provide an open-source repository for all the discussed methods, which is available at https://github.com/TaoWangzj/Awesome-Face-Restoration.

arXiv:2211.00754

BUbble Flow Field: a Simulation Framework for Evaluating Ultrasound Localization Microscopy Algorithms

Ultrasound contrast enhanced imaging has seen widespread uptake in research and clinical diagnostic imaging. This includes applications such as vector flow imaging, functional ultrasound and super-resolution Ultrasound Localization Microscopy (ULM). All of these require testing and validation during development of new algorithms with ground truth data. In this work we present a comprehensive simulation platform BUbble Flow Field (BUFF) that generates contrast enhanced ultrasound images in vascular tree geometries with realistic flow characteristics and validation algorithms for ULM. BUFF allows complex micro-vascular network generation of random and user-defined vascular networks. Blood flow is simulated with a fast Computational Fluid Dynamics (CFD) solver and allows arbitrary input and output positions and custom pressures. The acoustic field simulation is combined with non-linear Microbubble (MB) dynamics and simulates a range of point spread functions based on user-defined MB characteristics. The validation combines both binary and quantitative metrics. BFF's capacity to generate and validate user-defined networks is demonstrated through its implementation in the Ultrasound Localisation and TRacking Algorithms for Super Resolution (ULTRA-SR) Challenge at the International Ultrasonics Symposium (IUS) 2022 of the Institute of Electrical and Electronics Engineers (IEEE). The ability to produce ULM images, and the availability of a ground truth in localisation and tracking enables objective and quantitative evaluation of the large number of localisation and tracking algorithms developed in the field. BUFF can also benefit deep learning based methods by automatically generating datasets for training. BUFF is a fully comprehensive simulation platform for testing and validation of novel ULM techniques and is open source.