发布求助
文献互助 智能选刊 最新文献

2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)最新文献

筛选
英文 中文
Extended Guided Image Filtering For Contrast Enhancement 扩展引导图像滤波对比度增强
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9455989
Jiafei Wu, Genjie Li, Chong Wang, Huakai Liu, Shuai Zhang, Guangcheng Zhang
{"title":"Extended Guided Image Filtering For Contrast Enhancement","authors":"Jiafei Wu, Genjie Li, Chong Wang, Huakai Liu, Shuai Zhang, Guangcheng Zhang","doi":"10.1109/ICMEW53276.2021.9455989","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9455989","url":null,"abstract":"Guided image filtering (GIF) has been widely adopted in computer-vison tasks for properties of effective edge-preservation and low complexity. However, GIF is prone to suffer from halo artifacts near edges. Some new versions of GIF such as WGIF and GGIF employ various edge-aware regularizations to address this problem. But above local variance-based versions are sensitive to over-smoothing problem. The main reason is that local variance is not an adequate metric to discriminate the anisotropic image components. In this paper, we propose an extended GIF (EGIF) by incorporating a novel edge-aware factor into the regularization term. The new factor in entropy-variance domain not only can well discriminate details and edges but also have a good anti-noise ability. This helps EGIF to extend to industry applications. Experimental results of contrast enhancement show that EGIF can provide better visual quality as well as quantitative performance.","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127041583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Hierarchical Attention Image-Text Alignment Network For Person Re-Identification 用于人物再识别的分层关注图像-文本对齐网络
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9455960
Kajal Kansal, A. Subramanyam, Z. Wang, S. Satoh
{"title":"Hierarchical Attention Image-Text Alignment Network For Person Re-Identification","authors":"Kajal Kansal, A. Subramanyam, Z. Wang, S. Satoh","doi":"10.1109/ICMEW53276.2021.9455960","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9455960","url":null,"abstract":"Description based Person Re-identification (Re-ID) is a crucial cross-modality task that aims at retrieving a specific person for the given textual description. Existing description based Re-ID methods focus on learning robust representations to effectively measure the similarity between the global features of two modalities. However, such global mapping disregards semantic consistencies between local visual and linguistic features. Further, there are major challenges of alignment uncertainty that occur due to poor correspondence of text-image pairs and text complexity arising due to the irrelevant words. Towards this, we propose an end-to-end Hierarchical Attention Image-Text Alignment Network, named as HAITA-Net. Our model comprises of: i) a hierarchical attention alignment network to determine the potential relationships of image content and textual information at different levels, namely, word-patch level, phrase-patch level, and sentence-image level for addressing alignment uncertainty; ii) a new strategy of Term Frequency-Inverse document Frequency thresholding to extract the salient tokens to alleviate the challenge of text complexity. The network is optimized via joint weighted hierarchical attention loss and cross-modal loss in an end-to-end manner. Extensive experiments demonstrate the effectiveness of our method.","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116349385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
Cartoon Your Life 卡通你的生活
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9455964
Yezhi Shu, Ran Yi, Yong-Jin Liu
{"title":"Cartoon Your Life","authors":"Yezhi Shu, Ran Yi, Yong-Jin Liu","doi":"10.1109/ICMEW53276.2021.9455964","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9455964","url":null,"abstract":"The software transforms real-world photos into multi-style cartoons based on generative adversarial network (GAN). The web page of the source code is available 1, which is the implementation of our TVCG paper “GAN-based Multi-Style Photo Cartoonization” [1]. This code is also an improved version of our preliminary work on CVPR 2018 [2], which has drawn considerable attentions in image style translation area and so far has been cited 124 times in Google Scholar. A variant of these code have been deployed in Huawei Ascend chip processor, and incorporated into Atlas 200 DK Developer Kit and Atlas 300 Inference Server. Averagely 27k users per month can use our code to generate their own cartoons in Huawei Ascend community.","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125992008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
Attention-Based Encoder-Decoder Network For Single Image Dehazing 基于注意力的单幅图像去雾编码器网络
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9455979
Shunan Gao, Jinghua Zhu, Heran Xi
{"title":"Attention-Based Encoder-Decoder Network For Single Image Dehazing","authors":"Shunan Gao, Jinghua Zhu, Heran Xi","doi":"10.1109/ICMEW53276.2021.9455979","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9455979","url":null,"abstract":"Single image dehazing is a challenging problem in computer vision due to it is highly ill-posed. Although recent research has made great progress, the dehazed images produced by existing models still have residual haze and lose too much detail information. To solve the above problems, we propose an end-to-end Attention-based Encoder-Decoder Network (AEDNet) which is capable to effectively remove haze while preserving image details well. AEDNet employs a novel channel shuffle attention mechanism to adaptively adjust the weight of each channel-wise feature. This attention mechanism is integrated in residual block which is the core feature extraction module of encoder-decoder. Extensive experiments on synthetic datasets and real-hazy images demonstrate that our AEDNet achieves better performance compared with previous state-of-the-art methods.","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127470740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Cluster-Based Distribution Alignment For Generalizable Person Re-Identification 基于聚类的可泛化人员再识别分布对齐
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9455980
Chengzhang Zhu, Zhe Chang, Yalong Xiao, Beiji Zou, Bo Li, Shu Liu
{"title":"Cluster-Based Distribution Alignment For Generalizable Person Re-Identification","authors":"Chengzhang Zhu, Zhe Chang, Yalong Xiao, Beiji Zou, Bo Li, Shu Liu","doi":"10.1109/ICMEW53276.2021.9455980","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9455980","url":null,"abstract":"Person re-identification (ReID) aims to identify a person of interest across surveillance cameras. Although supervised ReID methods have made a great success, most of them suffer a large drop in performance on unseen domains. The key idea to solve this problem lies in the distribution alignment of pedestrian images. In this paper, we design a generalizable person ReID framework which is trained on the source domain but still generalizes well on unseen domains. To achieve this goal, our framework adaptively searches and aligns the underlying distributions among images through cluster-based Batch Normalization (BatchNorm). Specifically, the framework iteratively clusters images based on the style features extracted from middle layers of the networks. The clustering labels are then utilized to guide the cluster-based BatchNorm, which standardizes features in the same cluster within a batch. Extensive experiments on several ReID benchmarks demonstrate the effectiveness of our approach.","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130875180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
An Efficient Software for Building LIP Reading Models Without Pains 一个高效的软件，建立LIP阅读模型没有痛苦
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9456014
Dalu Feng, Shuang Yang, S. Shan
{"title":"An Efficient Software for Building LIP Reading Models Without Pains","authors":"Dalu Feng, Shuang Yang, S. Shan","doi":"10.1109/ICMEW53276.2021.9456014","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9456014","url":null,"abstract":"Lip reading is an impressive skill for human beings and has lots of potential applications. However, building an automatic lip-reading system is still a challenging task at present. Previous explorations of building a lip reading model usually contain several implicit operations and training details, which usually brings a big difficulty to the re-implementation for most researchers. In this work, we aim to establish a clear and efficient pipeline, which provides a convenient implementation tool and an easy start point for researchers and any others who want to study the problem of lip reading. We empirically study and introduce several useful training strategies in a clear and unified implementation procedure and compare their effects. Our pipeline got a performance of 88.4%/56.0% on two popular large-scale lipreading datasets, which is based on the basic model but achieves the performance comparable to or even higher than the state-of-the-art results. Our codes and models are available at https://github.com/Fengdalu/learn-aneffective-lip-reading-model-without-pains","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126674898","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 5
Practical Video Quality Assessment Of User Generated Content 用户生成内容的实用视频质量评估
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9455974
J. Korhonen, Xuanzheng Wen, Jun Cheng, Xu Wang
{"title":"Practical Video Quality Assessment Of User Generated Content","authors":"J. Korhonen, Xuanzheng Wen, Jun Cheng, Xu Wang","doi":"10.1109/ICMEW53276.2021.9455974","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9455974","url":null,"abstract":"During the past few years, video quality assessment (VQA) of user generated content (UGC) has attracted considerable attention in the research community. In this paper, we propose a practical architecture for a versatile video quality model, designed for assessing user generated videos in particular. The proposed architecture is based on our earlier design of two-level video quality model with a convolutional neural network (CNN-TLVQM), with various improvements and re-designed elements. We have built a fast implementation of the proposed model in C++, demonstrating that the model is practical for real-life applications. The implementation of the model has been submitted for evaluation in ICME UGCVQA Challenge in 2021.","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131128174","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Residual Proportion Multilayer Perceptron For Few-Shot Classification 残差比例多层感知器的少镜头分类
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9455947
Qifan Liu, Zhineng Zhao, W. Cao, Zhiquan He
{"title":"Residual Proportion Multilayer Perceptron For Few-Shot Classification","authors":"Qifan Liu, Zhineng Zhao, W. Cao, Zhiquan He","doi":"10.1109/ICMEW53276.2021.9455947","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9455947","url":null,"abstract":"Traditional deep learning methods require large amounts of data to ensure the excellent performance of the training model. The emergence of few-shot learning reduces the model’s dependence on data. Few-shot classification aims to identify novel categories with a small number of labeled samples. In this work, we tackle the few-shot classification task in the metric-based meta-learning method, which contains two parts: feature encoder and metric network. We used Graph Neural Network (GNN) as metric network and proposed RP-MLP (Residual Proportion MLP) to learn better the relationship among the samples. Our model is verified in two public datasets mini-ImageNet and tiered-ImageNet. We also conduct a cross-domain classification experiment on CUB200-2011. The results demonstrate that the method we proposed compares favorably against other state-of-the-art algorithms in few-shot classification tasks.","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131270499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Transformer Based Multi-Agent Framework 基于变压器的多代理框架
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9455984
Siyi Hu, Fengda Zhu, Xiaojun Chang, Xiaodan Liang
{"title":"Transformer Based Multi-Agent Framework","authors":"Siyi Hu, Fengda Zhu, Xiaojun Chang, Xiaodan Liang","doi":"10.1109/ICMEW53276.2021.9455984","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9455984","url":null,"abstract":"We present a Transformer-like agent to learn the policy of multi-agent cooperation tasks, which is a breakthrough for traditional RNN-based multi-agent models that need to be retrained for each task. Our model can handle various input and output with strong transferability and can parallel tackle different tasks. Besides, We are the first to successfully utilize transformer into a recurrent architecture, providing insight on stabilizing transformers in recurrent RL tasks.","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117060970","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Multi-Generator Adversarial Networks For Light Field Saliency Detection 用于光场显著性检测的多发生器对抗网络
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) Pub Date : 2021-07-05 DOI: 10.1109/ICMEW53276.2021.9456017
Hongyan Cai, Xudong Zhang, Rui Sun, R. Poppe, Jun Zhang
{"title":"Multi-Generator Adversarial Networks For Light Field Saliency Detection","authors":"Hongyan Cai, Xudong Zhang, Rui Sun, R. Poppe, Jun Zhang","doi":"10.1109/ICMEW53276.2021.9456017","DOIUrl":"https://doi.org/10.1109/ICMEW53276.2021.9456017","url":null,"abstract":"Compared to traditional cameras, light field imaging can additionally record the direction of incoming light. Based on this, light fields can be represented in focus stacks that focus at different depths and all-in-focus images. The refocusing information of focal stacks can provide supplementary information for saliency detection. In this paper, we propose a novel multi-generator adversarial network for saliency detection that consists of a cascaded multi-generator and a discriminator. The multi-generator extract saliency features from all-in-focus images and focal stacks. Besides, we set the predicted map multiplied by the all-in-focus image as the input of the discriminator. With this multiplication, we can preserve color and texture information of the salient objects, and reduce the computational cost. We conduct experiments on three public datasets. Compared with existing methods, our method achieves competitive results.","PeriodicalId":228350,"journal":{"name":"2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115866329","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 4
下一页 尾页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：604180095
Book学术官方微信
小红书