Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems最新文献_第2页

A Deep Learning model capable of producing heatmap probabilities for Characters in Natural Scenes. 一个深度学习模型，能够为自然场景中的角色生成热图概率。

Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems Pub Date : 2021-07-28 DOI: 10.1145/3480651.3480662

Allen Joshey, Ashish Tiwari, Rakesh Sankar, Sahil Salim Makandar

{"title":"A Deep Learning model capable of producing heatmap probabilities for Characters in Natural Scenes.","authors":"Allen Joshey, Ashish Tiwari, Rakesh Sankar, Sahil Salim Makandar","doi":"10.1145/3480651.3480662","DOIUrl":"https://doi.org/10.1145/3480651.3480662","url":null,"abstract":"Text appearing in Natural settings come in all shapes, sizes and textures. Classical methods have often failed at extracting accurately the text present in naturally occurring scenes. Text appearing in the wild presents itself in forms of hierarchy organized as sentences, words and characters. Methods for detecting Text from everyday scenes of the real world have found success. Most real world datasets available are annotated on a word level or line level thereby limiting detection to words and not characters. Inspired by the works of Naver Labs on CRAFT [2] and Microsoft Research and Baidu Research's work on WordSup [5] by training models in a weakly supervised manner to gain character level predictions. We propose a computationally efficient architecture capable of providing similar results. Thus our model, once capable of producing character level annotation trained on Synthetic text can be used to fine tune for text appearing in natural settings. The methods discussed prove to be robust enough to identify text that could be curved or somewhat deformed appearing in natural settings. Our approach includes the generation of probabilities of the location of characters and the gaps between characters of which constitute a word, such that it becomes easier to localize characters and words. Our method goes to show comparable results as to CRAFT [2] with only 30% of the number of learnable parameters required.","PeriodicalId":305943,"journal":{"name":"Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127743281","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

Simultaneous temporal and spatial deep attention for imaged skeleton-based action recognition 基于骨骼图像动作识别的同时时间和空间深度注意

Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems Pub Date : 2021-07-28 DOI: 10.1145/3480651.3480668

Mohamed Lamine Rouali, Said Yacine Boulahia, Abdenour Amamra

{"title":"Simultaneous temporal and spatial deep attention for imaged skeleton-based action recognition","authors":"Mohamed Lamine Rouali, Said Yacine Boulahia, Abdenour Amamra","doi":"10.1145/3480651.3480668","DOIUrl":"https://doi.org/10.1145/3480651.3480668","url":null,"abstract":"The use of skeletons as a modality to represent and recognize human actions has gained interest thanks to the compactness of the data, their reliable representativeness in addition to their strong robustness. The deep learning based recognition approaches which are based on it often propose to improve the recognition pipeline by integrating the concept of attention in their modeling. The idea is to allow the model to focus on the relevant information of the action instead of attempting some kind of blind modeling. In this article, we propose an action recognition approach integrating simultaneously both spatial and temporal attentions. We first perform a transformation of the input sequence data into a color matrix, called imaged skeleton, comprising Cartesian and rotational information. Then, this new representation is given as input to an architecture composed of a main trunk, that allows features extraction and classification, and several attention branches. Different experimental evaluations on two popular benchmark databases, namely UT-Kinect [1] and SBU Kinect Interaction [2], are conducted to verify the interest of our proposed approach, where better performances are reported. Index: convolutional neural network, spatio-temporal, skeleton-based action recognition, deep attention.","PeriodicalId":305943,"journal":{"name":"Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121199136","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

Cell Detection by Robust Self-Trained Networks 基于鲁棒自训练网络的细胞检测

Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems Pub Date : 2021-07-28 DOI: 10.1145/3480651.3480665

Yuang Zhu, Yuxin Zheng, Zhao Chen

引用次数: 0

Face anti-spoofing by using Feature Fusion 利用特征融合实现抗欺骗

Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems Pub Date : 2021-07-28 DOI: 10.1145/3480651.3480658

Qiong Liu, Lan Zhang

引用次数: 0

Food safety pre-warning system based on Robust Principal Component Analysis and Improved Apriori Algorithm 基于鲁棒主成分分析和改进Apriori算法的食品安全预警系统

Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems Pub Date : 2021-07-28 DOI: 10.1145/3480651.3480653

Xiaowen Ding, Sheng Xu

引用次数: 2

Age Estimation from Facial Images using Transfer Learning and K-fold Cross-Validation 使用迁移学习和K-fold交叉验证的面部图像年龄估计

Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems Pub Date : 2021-07-28 DOI: 10.1145/3480651.3480659

S. M. S. Uddin, Md. Samin Morshed, Mahruf Islam Prottoy, A. Rahman

引用次数: 4

Application of Wavelet Analysis in Image Matching 小波分析在图像匹配中的应用

Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems Pub Date : 2021-07-28 DOI: 10.1145/3480651.3480670

Linglong Tan, Fengzhi Wu, Xiaoyao Yin, Song Xue

{"title":"Application of Wavelet Analysis in Image Matching","authors":"Linglong Tan, Fengzhi Wu, Xiaoyao Yin, Song Xue","doi":"10.1145/3480651.3480670","DOIUrl":"https://doi.org/10.1145/3480651.3480670","url":null,"abstract":"Abstract. Based on the study of traditional matching methods, this paper implements a low-frequency image matching system based on wavelet transform, which is composed of wavelet preprocessing, low-frequency image extraction, and image matching. The low-frequency image after wavelet decomposition is used for matching, which can reduce the calculation time of matching. The low-frequency image still contains most of the visual information of the original image, making the matching result stable and reliable.In this system, image wavelet decomposition and matching use mature and fast algorithms. The matching is performed on low-frequency images, which makes the amount of calculation for matching very small. Using the low-frequency components of the image to match also greatly removes the interference of noise on the image matching. Since the highest proportion of high-frequency noise in the noise has been removed before the algorithm is matched, all the matching algorithms have good anti-noise ability.The matching system in this paper adopts a matching method based on low-frequency components after wavelet transform, discusses and realizes the use of low-frequency images after image wavelet decomposition to perform image matching. The experimental results show that the matching algorithm used in the article has fast calculation speed, less matching time, and certain practicability.","PeriodicalId":305943,"journal":{"name":"Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133317960","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

Design and Implementation of the Wood Knot Recognition System Based on Matlab GUI 基于Matlab GUI的木结识别系统的设计与实现

Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems Pub Date : 2021-07-28 DOI: 10.1145/3480651.3480664

Xiaoxia Yang, Xin Gao, Tao Wang, Kepeng Yang, Chengxin Hu, Xiaoping Liu, Yucheng Zhou

引用次数: 0

Synthetic Aperture Radar image target recognition based on hybrid attention mechanism 基于混合注意机制的合成孔径雷达图像目标识别

Proceedings of the 2021 International Conference on Pattern Recognition and Intelligent Systems Pub Date : 2021-07-28 DOI: 10.1145/3480651.3480660

Baodai Shi, Qin Zhang, Yao Li

引用次数: 0