220 Hits in 8.2 sec

MC-Blur: A Comprehensive Benchmark for Image Deblurring [article]

Kaihao Zhang, Tao Wang, Wenhan Luo, Boheng Chen, Wenqi Ren, Bjorn Stenger, Wei Liu, Hongdong Li, Ming-Hsuan Yang
2022 arXiv   pre-print
For in-depth performance evaluation, we construct a new large-scale multi-cause image deblurring dataset (called MC-Blur), including real-world and synthesized blurry images with mixed factors of blurs  ...  These benchmarking results provide a comprehensive overview of the advantages and limitations of current deblurring methods, and reveal the advances of our dataset.  ...  To address these problems, we propose a comprehensive and large-scale multi-cause image deblurring dataset called MC-Blur.  ... 
arXiv:2112.00234v2 fatcat:my6fu4g2yrf3hiyn56wnv2omvq

LODE: Deep Local Deblurring and A New Benchmark [article]

Zerun Wang, Liuyu Xiang, Fan Yang, Jinzhao Qian, Jie Hu, Haidong Huang, Jungong Han, Yuchen Guo, Guiguang Ding
2021 arXiv   pre-print
In this paper, we first lay the data foundation for local deblurring by constructing, for the first time, a LOcal-DEblur (LODE) dataset consisting of 3,700 real-world captured locally blurred images and  ...  While recent deep deblurring algorithms have achieved remarkable progress, most existing methods focus on the global deblurring problem, where the image blur mostly arises from severe camera shake.  ...  To our best knowledge, LODE is the first benchmark dataset for the local deblurring task. 3. We propose a novel BladeNet for deep local deblurring.  ... 
arXiv:2109.09149v1 fatcat:ckwicbqeavglldloemr7smsbxq

UG^2: A Video Benchmark for Assessing the Impact of Image Restoration and Enhancement on Automatic Visual Recognition

Rosaura G. Vidal, Sreya Banerjee, Klemen Grm, Vitomir Struc, Walter J. Scheirer
2018 2018 IEEE Winter Conference on Applications of Computer Vision (WACV)  
To facilitate new research, we introduce a new benchmark dataset called UG^2, which contains three difficult real-world scenarios: uncontrolled videos taken by UAVs and manned gliders, as well as controlled  ...  Results showthat there is plenty of room for algorithmic innovation, making this dataset a useful tool going forward.  ...  Adam Czajka, visiting assistant professor at the University of Notre Dame and Mr. Sebastian Kawa for assistance with data collection.  ... 
doi:10.1109/wacv.2018.00177 dblp:conf/wacv/VidalBGSS18 fatcat:whgwndjdorfx7mijj6t4nv7ugy

Learning Enriched Features for Fast Image Restoration and Enhancement [article]

Syed Waqas Zamir, Aditya Arora, Salman Khan, Munawar Hayat, Fahad Shahbaz Khan, Ming-Hsuan Yang, Ling Shao
2022 arXiv   pre-print
The core of our approach is a multi-scale residual block containing the following key elements: (a) parallel multi-resolution convolution streams for extracting multi-scale features, (b) information exchange  ...  deblurring, image denoising, super-resolution, and image enhancement.  ...  Recent advances in image restoration and enhancement have been led by deep learning models, as they can learn strong (generalizable) priors from large-scale datasets.  ... 
arXiv:2205.01649v1 fatcat:hagwkoo5b5dahkv5apc5lmwtom

Two-level Wavelet-based Convolutional Neural Network for Image Deblurring

Yeyun Wu, Pan Qian, Xiaofeng Zhang
2021 IEEE Access  
This is a benchmark dataset widely used for single image deblurring algorithms. Our model is also tested on this dataset. A.  ...  The model has three scales and a total of 120 ResBlocks, which causes too many network parameters. Zhang et al.  ... 
doi:10.1109/access.2021.3067055 fatcat:5hq3dsu3ercjffmi3dib3h2gzy

Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model [article]

Zhiyuan Mao and Ajay Jaiswal and Zhangyang Wang and Stanley H. Chan
2022 arXiv   pre-print
In addition, recognizing the lack of a comprehensive dataset, we collect and present two new real-world turbulence datasets that allow for evaluation with both classical objective metrics (e.g., PSNR and  ...  SSIM) and a new task-driven metric using text recognition accuracy.  ...  Foundation under the grants CCSS-2030570 and IIS-2133032.  ... 
arXiv:2207.10040v2 fatcat:trghxfowhne3vjz3n2nlhzq5fq

Human-Aware Motion Deblurring [article]

Ziyi Shen, Wenguan Wang, Xiankai Lu, Jianbing Shen, Haibin Ling, Tingfa Xu, Ling Shao
2020 arXiv   pre-print
To further benefit the research towards Human-aware Image Deblurring, we introduce a large-scale dataset, named HIDE, which consists of 8,422 blurry and sharp image pairs with 65,784 densely annotated  ...  Extensive experiments on public benchmarks and our dataset demonstrate that our model performs favorably against the state-of-the-art motion deblurring methods, especially in capturing semantic details  ...  is explored to explicitly model FG/BG motion blur and comprehensively fuse different-domain information for global and harmonious deblurring. • A large-scale dataset, HIDE, is carefully constructed for  ... 
arXiv:2001.06816v1 fatcat:iihosraewzhahebknotnrtzcwq

Deep Deblurring Correlation Filter for Object Tracking

Yu Bai, Tingfa Xu, Bo Huang, Ruoling Yang
2020 IEEE Access  
We employ RNN as our baseline of deblurnet, and introduce residual block and ConvLSTM in our deblur network to improve the result of deblurring.  ...  Extensive experimental results demonstrate our tracker outperforms several state-of-the-art trackers on the OTB-2015 and VOT-2016 datasets.  ...  AUTHOR CONTRIBUTIONS Yu Bai and Tingfa Xu designed the global structure and the experiments. Ruoling Yang and Yu Bai performed software experiments and analyzed the data. Yu Bai wrote the paper.  ... 
doi:10.1109/access.2020.2986311 fatcat:oca3qa2gsrdybbtgtund4352r4

Motion Deblurring with an Adaptive Network [article]

Kuldeep Purohit, A. N. Rajagopalan
2022 arXiv   pre-print
Our networks can implicitly model the spatially-varying deblurring process, while dispensing with multi-scale processing and large filters entirely.  ...  Extensive qualitative and quantitative comparisons with prior art on benchmark dynamic scene deblurring datasets clearly demonstrate the superiority of the proposed networks via reduction in model-size  ...  The methods [28, 50] use a multi-scale strategy to improve capability to handle large blur, but fail in challenging situations.  ... 
arXiv:1903.11394v4 fatcat:jgsssxep6vfbnafshrg2x6ijke

A Lightweight Fusion Distillation Network for Image Deblurring and Deraining

Yanni Zhang, Yiming Liu, Qiang Li, Jianzhong Wang, Miao Qi, Hui Sun, Hui Xu, Jun Kong
2021 Sensors  
In the encoding stage, the image feature is reduced to various small-scale spaces for multi-scale information extraction and fusion without much information loss.  ...  Recently, deep learning-based image deblurring and deraining have been well developed. However, most of these methods fail to distill the useful features.  ...  [12] investigated a new scheme that exploits the deblurring cues at different scales via a hierarchical multi-patch model and proposed a simple yet effective multi-level CNNs model called Deep Multi-Patch  ... 
doi:10.3390/s21165312 pmid:34450762 pmcid:PMC8398398 fatcat:uilblt4l45bmxp4oe4tt4lm7dq

MBA-VO: Motion Blur Aware Visual Odometry [article]

Peidong Liu, Xingxing Zuo, Viktor Larsson, Marc Pollefeys
2021 arXiv   pre-print
In addition, we also contribute a novel benchmarking dataset for motion blur aware visual odometry.  ...  In this paper we present a novel hybrid visual odometry pipeline with direct approach that explicitly models and estimates the camera's local trajectory within the exposure time.  ...  [14] later propose a 15 layer network for text image deblurring. The network was further enlarged to 40 layers in a multi-scale manner by Nah et al.  ... 
arXiv:2103.13684v1 fatcat:zwdcusorprd77j34nji6uupy7u

Multi-Scale Neural Network with Dilated Convolutions for Image Deblurring

Jose Jaena Mari Ople, Pin-Yi Yeh, Shih-Wei Sun, I-Te Tsai, Kai-Lung Hua
2020 IEEE Access  
For example, large blurs, such as those caused by fast-moving objects leaving a trail of afterimages, need spatial context from a large region, while small blurs, such as those caused by slight camera  ...  of the images, and (2) dilated convolutions that can extract multi-scale features by using different dilation rates.  ...  Our proposed model can restore blur, caused by complex or large movement, with excellent results and less visual artifacts.  ... 
doi:10.1109/access.2020.2980996 fatcat:qbgmxfawrbf4palcfqj7lhqwtq

High-Resolution Dual-Stage Multi-Level Feature Aggregation for Single Image and Video Deblurring

Stephan Brehm, Sebastian Scherer, Rainer Lienhart
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
We present a model that combines highresolution processing with a multi-resolution feature aggregation method for single frame and video deblurring. Our proposed model consists of 2 stages.  ...  We apply our framework on current benchmarks and challenges and show that our model provides state-of-the art results.  ...  We report results on the validation set and test set if available. We further trained our models on the GoPro dataset [24] , which is a standard benchmark for evaluation of deblurring algorithms.  ... 
doi:10.1109/cvprw50498.2020.00237 dblp:conf/cvpr/BrehmSL20 fatcat:zugonisrobegrgroarq2hf7g6y

Region-Adaptive Dense Network for Efficient Motion Deblurring

Kuldeep Purohit, A. N. Rajagopalan
Our network facilitates interpretable modeling of the spatially-varying deblurring process while dispensing with multi-scale processing and large filters entirely.  ...  Extensive comparisons with prior art on benchmark dynamic scene deblurring datasets clearly demonstrate the superiority of the proposed networks via significant improvements in accuracy and speed, enabling  ...  Till date, the driving force behind performance improvement in deblurring has been use of large number of layers, larger filters, and multi-scale processing which assist in increasing the "static" receptive  ... 
doi:10.1609/aaai.v34i07.6862 fatcat:pu6mz3jasfexnidcg3rawsrf5q

Report on UG^2+ Challenge Track 1: Assessing Algorithms to Improve Video Object Detection and Classification from Unconstrained Mobility Platforms [article]

Sreya Banerjee, Rosaura G. VidalMata, Zhangyang Wang, Walter J. Scheirer
2020 arXiv   pre-print
The challenge made use of the UG^2 (UAV, Glider, Ground) dataset, which is an established benchmark for assessing the interplay between image restoration and enhancement and visual recognition. 16 algorithms  ...  One promising option is to make use of image restoration and enhancement algorithms from the area of computational photography to improve the quality of the underlying frames in a way that also improves  ...  Datasets with a similar type of data to the one employed for this challenge include large-scale video surveillance datasets such as [19, 20, 21, 22] , which provide video captured from a single fixed  ... 
arXiv:1907.11529v4 fatcat:euvqbhquejfv7ma366wxzc3xtq
« Previous Showing results 1 — 15 out of 220 results