检索结果-内蒙古大学图书馆

2017 ieee visual communications and image processing, vcip 2017

ISBN: (纸本)9781538604625

The proceedings contain 142 papers. The topics discussed include: perceptual optimized adaptive HTTP streaming;TC-CCS: a cooperative caching strategy in mobile transparent computing system;non-invasive imaging based on speckle pattern estimation and deconvolution;object localization in weakly labeled data using regularized attention networks;dictionary learning based high frequency inter-layer prediction for scalable HEVC;CNN oriented fast QTBT partition algorithm for JVET intra coding;endoscopic video deblurring via synthesis;improve object detection via a multi-feature and multi-task CNN model;enhancing pulmonary nodule detection via cross-modal alignment;adaptive difference modelling for background subtraction;geometry-adaptive motion partitioning using improved temporal prediction;cascaded convolutional neural networks for object detection;a new underwater image enhancing method via color correction and illumination adjustment;a novel image compression framework at edges;QoE-based node selection strategy for edge computing enabled internet-of-vehicles (EC-IoV);deep probabilities for age estimation;and deep discriminative network with inception module for person re-identification.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Practical Approach to Depth-Aware Augmentation for Neural Radiance Fields

A Practical Approach to Depth-Aware Augmentation for Neural ...

引用

2024 Conference on visual communications and image processing

作者： Khosroshahi, Hamed Razavi Sancho, Jaime Bonatto, Daniele Fachada, Sarah Bang, Gun Lafruit, Gauthier Juarez, Eduardo Teratani, Mehrdad Univ Libre Bruxelles Lab Image Synth & Anal Brussels Belgium Univ Politecn Madrid Res Ctr Software Technol & Multimedia Syst Madrid Spain Telecommun Res Inst Elect Daejeon South Korea

ISBN: (纸本)9798331529543;9798331529550

Neural Radiance Fields (NeRF) have demonstrated exceptional performance in generating novel views of scenes by learning implicit volumetric representations from calibrated RGB images, without depth information. A major limitation is the need for large training datasets in neural network-based view synthesis frameworks. The challenge of effective data augmentation for view synthesis remains unresolved. NeRF models require extensive scene coverage from multiple views to accurately estimate radiance and density. Insufficient coverage reduces the model's ability to interpolate or extrapolate unseen parts of the scene effectively. In this paper, we propose a novel pipeline to address this data augmentation issue using depth map information. We use depth image-based rendering (DIBR) to overcome the lack of enough views for training NeRF. Experimental results indicate that our approach enhances the quality of rendered images using the NeRF framework, achieving an average peak signal-to-noise ratio (PSNR) increase of 7.2 dB, with a maximum improvement of 12 dB.

关键词： Neural Radiance Fields NeRF View synthesis Data augmentation Depth map

来源：评论

学校读者我要写书评

暂无评论

Adaptive Threshold-based Sparse Representation Network for image Compressive Sensing Reconstruction

Adaptive Threshold-based Sparse Representation Network for I...

引用

ieee International Conference on visual communications and image processing (vcip) - visual communications in the Era of AI and Limited Resources

作者： Xuan, Yunyi Yang, Chunling Yang, Xin South China Univ Technol Sch Elect & Informat Engn Guangzhou Peoples R China

ISBN: (纸本)9781728185514

Recently, network-based image Compressive Sensing (ICS) algorithms show superior performance in reconstruction quality and speed, yet non-interpretable. Herein, we propose an Adaptive Threshold-based Sparse Representation Reconstruction Network (ATSR-Net), composed of the Convolutional Sparse Representation subnet (CSR-subnet) and the truly Adaptive Threshold Generation subnet (ATG-subnet). The traditional iterations are unfolded into several CSR-subnets, which can fully exploit the local and nonlocal similarities. The ATG-subnet automatically determines a threshold map based on the image intrinsic characterization for flexible feature selection. Moreover, we present a three-level consistency loss based on pixel-level, measurement-level, and feature-level, to accelerate the network convergence. Extensive experiment results demonstrate the superiority of the proposed network to the existing state-of-the-art methods by large margins, both quantitatively and qualitatively.

关键词： image compressive sensing sparse representation adaptive threshold generation

来源：评论

学校读者我要写书评

暂无评论

Robust Perturbation for visual Explanation: Cross-Checking Mask Optimization to Avoid Class Distortion

引用

ieee TRANSACTIONS ON image processing 2022年 31卷 301-313页

作者： Kim, Junho Kim, Seongyeop Kim, Seong Tae Ro, Yong Man Korea Adv Inst Sci & Technol KAIST Sch Elect Engn Image & Video Syst Lab Daejeon 34141 South Korea Kyung Hee Univ Dept Comp Sci & Engn Yongin 17104 Gyeonggi Do South Korea

Along with the outstanding performance of the deep neural networks (DNNs), considerable research efforts have been devoted to finding ways to understand the decision of DNNs structures. In the computer vision domain, visualizing the attribution map is one of the most intuitive and understandable ways to achieve human-level interpretation. Among them, perturbation-based visualization can explain the "black box" property of the given network by optimizing perturbation masks that alter the network prediction of the target class the most. However, existing perturbation methods could make unexpected changes to network predictions after applying a perturbation mask to the input image, resulting in a loss of robustness and fidelity of the perturbation mechanisms. In this paper, we define class distortion as the unexpected changes of the network prediction during the perturbation process. To handle that, we propose a novel visual interpretation framework, Robust Perturbation, which shows robustness against the unexpected class distortion during the mask optimization. With a new cross-checking mask optimization strategy, our proposed framework perturbs the target prediction of the network while upholding the non-target predictions, providing more reliable and accurate visual explanations. We evaluate our framework on three different public datasets through extensive experiments. Furthermore, we propose a new metric for class distortion evaluation. In both quantitative and qualitative experiments, tackling the class distortion problem turns out to enhance the quality and fidelity of the visual explanation in comparison with the existing perturbation-based methods.

关键词： Distortion Perturbation methods visualization Optimization Cats Bicycles Automobiles visual explanation attribution map class distortion mask perturbation

来源：评论

学校读者我要写书评

暂无评论

Semantic Compensation Based Dual-Stream Feature Interaction Network for Multi-oriented Scene Text Detection

Semantic Compensation Based Dual-Stream Feature Interaction ...

引用

ieee visual communications and image processing (vcip)

作者： Siyan Wang Sumei Li School of Electrical and Information Engineering Tianjin University Tianjin China

ISBN: (纸本)9781665475938

Due to the various appearances of scene text instances and the disturbance of background, it is still a challenging task to design an effective and accurate text detector. To tackle this problem, in this paper we propose a novel dual-stream scene text detector considering semantic compensation and feature interaction. The detector extracts image features from two input images of different resolution, which improves its perceptive ability and contributes to detecting large and long texts. Specifically, we propose a Semantic Compensation Module (SCM) to aggregate features between the two streams, which compensates semantic information in features at each level via an attention mechanism. Moreover, we design a Feature Interaction Module (FIM) to obtain more expressive features. Experiments conducted on three benchmark datasets, ICDAR2015, MSRA-TD500 and ICDAR2017-MLT, demonstrate that our proposed method has competitive performance and strong robustness.

关键词： image resolution visual communication Semantics Neural networks Detectors Streaming media Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Weight-Based Mask for Domain Adaptation 48

Weight-Based Mask for Domain Adaptation

引用

48th ieee International Conference on Acoustics, Speech and Signal processing, ICASSP 2023

作者： Lee, Eunseop Kim, Inhan Kim, Daijin Pohang University of Science and Technology Department of Computer Science and Engineering Korea Republic of

ISBN: (纸本)9781728163277

In computer vision, unsupervised domain adaptation (UDA) is an approach to transferring knowledge from a label-rich source domain to a fully-unlabeled target domain. Conventional UDA approaches have two problems. The first problem is that a class classifier can be biased to the source domain because it is trained using only source samples. The second is that previous approaches align image-level features regardless of foreground and background, although the classifier requires foreground features. To solve these problems, we introduce Weight-based Mask Network (WEMNet) composed of Domain Ignore Module (DIM) and Semantic Enhancement Module (SEM). DIM obtains domain-Agnostic feature representations via the weight of the domain discriminator and predicts categories. In addition, SEM obtains class-related feature representations using the classifier weight and focuses on the foreground features for domain adaptation. Extensive experimental results reveal that the proposed WEMNet outperforms the competitive accuracy on representative UDA datasets. © 2023 ieee.

关键词： image classification

来源：评论

学校读者我要写书评

暂无评论

Video Quality Model of Compression, Resolution and Frame Rate Adaptation Based on Space-Time Regularities

引用

ieee TRANSACTIONS ON image processing 2022年 31卷 3644-3656页

作者： Lee, Dae Yeol Kim, Jongho Ko, Hyunsuk Bovik, Alan C. Univ Texas Austin Dept Elect & Comp Engn Austin TX 78712 USA ETRI Media Coding Res Sect Daejeon 34129 South Korea Hanyang Univ ERICA Sch Elect Engn Ansan 15588 South Korea

Being able to accurately predict the visual quality of videos subjected to various combinations of dimension reduction protocols is of high interest to the streaming video industry, given rapid increases in frame resolutions and frame rates. In this direction, we have developed a video quality predictor that is sensitive to spatial, temporal, or space-time subsampling combined with compression. Our predictor is based on new models of space-time natural video statistics (NVS). Specifically, we model the statistics of divisively normalized difference between neighboring frames that are relatively displaced. In an extensive empirical study, we found that those paths of space-time displaced frame differences that provide maximal regularity against our NVS model generally align best with motion trajectories. Motivated by this, we built a new video quality prediction engine that extracts NVS features that represent how space-time directional regularities are disturbed by space-time distortions. Based on parametric models of these regularities, we compute features that are used to train a regressor that can accurately predict perceptual quality. As a stringent test of the new model, we apply it to the difficult problem of predicting the quality of videos subjected not only to compression, but also to downsampling in space and/or time. We show that the new quality model achieves state-of-the-art (SOTA) prediction performance on the new ETRI-LIVE Space-Time Subsampled Video Quality (STSVQ) and also on the AVT-VQDB-UHD-1 database.

关键词： Streaming media Video recording Quality assessment Predictive models image coding Distortion Video compression Video quality natural video statistics statistical regularity space-time displaced frame differences space-time resolution video compression

来源：评论

学校读者我要写书评

暂无评论

Multi-Dimension Aware Back Projection Network For Scene Text Detection

Multi-Dimension Aware Back Projection Network For Scene Text...

引用

ieee visual communications and image processing (vcip)

作者： Yizhan Zhao Sumei Li Yongli Chang School of Electrical and Information Engineering Tianjin University Tianjin China

ISBN: (纸本)9781728173221

Recently, scene text detection based on deep learning has progressed substantially. Nevertheless, most previous models with FPN are limited by the drawback of sample interpolation algorithms, which fail to generate high-quality up-sampled features. Accordingly, we propose an end-to-end trainable text detector to alleviate the above dilemma. Specifically, a Back Projection Enhanced Up-sampling (BPEU) block is proposed to alleviate the drawback of sample interpolation algorithms. It significantly enhances the quality of up-sampled features by employing back projection and detail compensation. Further-more, a Multi-Dimensional Attention (MDA) block is devised to learn different knowledge from spatial and channel dimensions, which intelligently selects features to generate more discriminative representations. Experimental results on three benchmarks, ICDAR2015, ICDAR2017- MLT and MSRA-TD500, demonstrate the effectiveness of our method.

关键词： Deep learning Interpolation visual communication Shape image processing Detectors Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Pseudo-Supervised Learning for Semantic Multi-Style Transfer

引用

ieee ACCESS 2021年 9卷 7930-7942页

作者： Kim, Saehun Do, Jeonghyeok Kim, Munchurl Korea Adv Inst Sci & Technol Sch Elect Engn Daejeon 34141 South Korea

Numerous methods for style transfer have been developed using unsupervised learning and gained impressive results. However, optimal style transfer cannot be conducted from a global fashion in certain style domains, mainly when a single target-style domain contains semantic objects that have their own distinct and unique styles, e.g., those objects in the anime-style domain. Previous methods are incongruent because the unsupervised learning can not provide the semantic mappings between the multi-style objects according to their unique styles. Thus, in this paper, we propose a pseudo-supervised learning framework for the semantic multi-style transfer (SMST), which consists of (i) a pseudo ground truth (pGT) generation phase and (ii) a SMST learning phase. In the pGT generation phase, multiple semantic objects of the photo images are separately transferred to the target-domain object styles in an object-oriented fashion. Then the transferred objects are composed back to an image, which is the pGT. In the SMST learning phase, a SMST network (SMSTnet) is trained with the pairs of the photo images and its respective pGT in a supervised manner. From this, our framework can provide the semantic mappings of multi-style objects. Moreover, to embrace the multi-styles of various objects into a single generator, we design the SMSTnet with channel attentions in conjunction with a discriminator dedicated to our pseudo-supervised learning. Our method has been applied and intensively tested for anime-style transfer learning. The experimental results demonstrate the effectiveness of our method and show its superiority compared to the state-of-the-art methods.

关键词： Style transfer image-to-image translation generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

SdBAN: Salient Object Detection Using Bilateral Attention Network With Dice Coefficient Loss

引用

ieee ACCESS 2020年 8卷 104357-104370页

作者： Kang, Donggoo Park, Sangwoo Paik, Joonki Chung Ang Univ Dept Image Seoul 06974 South Korea

visual attention plays an important role in saliency detection by highlighting meaningful context regions. In this paper, we present a novel saliency detection method using a bilateral attention network. The proposed network consists of two branches: i) a spatial path using an encoder-decoder structure to learn spatial cues and ii) a context path using an attention mechanism to learn contextual cues. The feature aggregation module is finally used to predict salient objects by concatenating the cues. To optimize the weights of the network in the sense of minimizing the class imbalance problem, we minimize the dice coefficient loss together with the classical cross-entropy loss. The proposed network can predict salient regions in an end-to-end manner without post-processing. Experimental results show that the proposed network achieved better performance than existing state-of-the-art methods in most cases. Furthermore, the proposed network takes only 0.03 seconds to process a image. The code for the proposed method can be found at the following

关键词： Salient object detection deep learning dice coefficient attention mechanism

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：