计算机视觉与模式识别学术速递[2022.4.25]

 重大更新！公众号每日速递覆盖arXiv所有方向,涵盖CS|物理|数学|经济|统计|金融|生物|电气等领域。点击查看历史文章列表，可在公众号内订阅#arXiv每日学术速递话题，接收每日推送。点击访问arxivdaily.com每日更新更及时。

计算机视觉与模式识别cs.CV 方向，今日共计52篇

【Transformer】(4篇)

【1】 ParkPredict+: Multimodal Intent and Motion Prediction for Vehicles in Parking Lots with CNN and Transformer

标题：ParkPredict+：使用CNN和Transformer对停车场车辆的多模式意图和运动预测

作者：Xu Shen,Matthew Lacayo,Nidhir Guggilla,Francesco Borrelli

机构*： TheirUniversity of California

链接：点击下载PDF文件

【2】 Diverse Instance Discovery: Vision-Transformer for Instance-Aware Multi-Label Image Recognition

标题：多样性实例发现：实例感知多标签图像识别的视觉转换器

作者：Yunqing Hu,Xuan Jin,Yin Zhang,Haiwen Hong,Jingfeng Zhang,Feihu Yan,Yuan He,Hui Xue

机构*：College of Computer Science and Technology, Zhejiang University, Hangzhou, China, Alibaba Group, Hangzhou, China

备注：Accepted to ICME 2022

链接：点击下载PDF文件

【3】 Spatiality-guided Transformer for 3D Dense Captioning on Point Clouds

标题：基于空间制导的点云三维密集字幕变换

作者：Heng Wang,Chaoyi Zhang,Jianhui Yu,Weidong Cai

机构*：School of Computer Science, University of Sydney, Australia

备注：IJCAI 2022. Project Page: this https URL

链接：点击下载PDF文件

【4】 Hypergraph Transformer: Weakly-supervised Multi-hop Reasoning for Knowledge-based Visual Question Answering

标题：超图变换：基于知识的视觉问答的弱监督多跳推理

作者：Yu-Jung Heo,Eun-Sol Kim,Woo Suk Choi,Byoung-Tak Zhang

机构*： Seoul National University , Department of Computer Science, Hanyang University, AI Institute (AIIS), Seoul National University , Surromind

备注：Accepted at ACL 2022

链接：点击下载PDF文件

【检测相关】(5篇)

【1】 Pay "Attention" to Adverse Weather: Weather-aware Attention-based Object Detection

作者：Saket S. Chaturvedi,Lan Zhang,Xiaoyong Yuan

机构*：College of Computing, Michigan Technological University, Houghton, MI, USA, Dept of Electrical and Computer Engineering

备注：This paper is accepted at IEEE International Conference on Pattern Recognition (ICPR), 2022

链接：点击下载PDF文件

【2】 Unknown Face Presentation Attack Detection via Localised Learning of Multiple Kernels

标题：基于多核局部学习的未知人脸呈现攻击检测

作者：Shervin Rahimzadeh Arashloo

机构*： University of Surrey

链接：点击下载PDF文件

【3】 DFAM-DETR: Deformable feature based attention mechanism DETR on slender object detection

标题：DFAM-DETR：基于变形特征的细长目标检测注意机制

作者：Wen Feng,Wang Mei,Hu Xiaojie

机构*：Shenyang Ligong University

备注：9 pages, 7 figures

链接：点击下载PDF文件

【4】 Real-time HOG+SVM based object detection using SoC FPGA for a UHD video stream

标题：基于SoC FPGA的超高清视频流实时HOG+支持向量机目标检测

作者：Mateusz Wasala,Tomasz Kryjak

机构*：Department of Automatic Control and Robotics, AGH University of Science and Technology, Krakow, Poland

备注：6 pages, accepted for the CPS & IoT 2022 conference

链接：点击下载PDF文件

【5】 Few-Shot Object Detection with Proposal Balance Refinement

标题：基于提案平衡细化的Few-Shot目标检测

作者：Sueyeon Kim,Woo-Jeoung Nam,Seong-Whan Lee

机构*：. Department of Artificial Intelligence, Korea University, . Department of Computer and Radio Communications Engineering, Korea University

链接：点击下载PDF文件

【分类|识别相关】(2篇)

【1】 iCAR: Bridging Image Classification and Image-text Alignment for Visual Recognition

标题：ICAR：用于视觉识别的图像分类和图文对齐的桥梁

作者：Yixuan Wei,Yue Cao,Zheng Zhang,Zhuliang Yao,Zhenda Xie,Han Hu,Baining Guo

机构*：Tsinghua University, Microsoft Research Asia

备注：22 pages, 6 figures

链接：点击下载PDF文件

【2】 Reinforcing Generated Images via Meta-learning for One-Shot Fine-Grained Visual Recognition

标题：基于元学习的单次细粒度视觉识别增强生成图像

作者：Satoshi Tsutsui,Yanwei Fu,David Crandall

机构*： Fudan University, ZhejiangNormal University

备注：Accepted to PAMI 2022. arXiv admin note: substantial text overlap with arXiv:1911.07164

链接：点击下载PDF文件

【分割|语义相关】(6篇)

【1】 Self-Supervised Video Object Segmentation via Cutout Prediction and Tagging

标题：基于剪裁预测和标注的自监督视频对象分割

作者：Jyoti Kini,Fahad Shahbaz Khan,Salman Khan,Mubarak Shah

机构*：University of Central Florida, USA, MBZ University of Artificial Intelligence, Linköping University, UAE, Sweden

链接：点击下载PDF文件

【2】 Tag-Based Attention Guided Bottom-Up Approach for Video Instance Segmentation

标题：基于标签的注意力引导自下而上的视频实例分割方法

作者：Jyoti Kini,Mubarak Shah

机构*：Center for Research in Computer Vision, University of Central Florida, Orlando, Florida

链接：点击下载PDF文件

【3】 EmbedTrack -- Simultaneous Cell Segmentation and Tracking Through Learning Offsets and Clustering Bandwidths

标题：EmbedTrack--基于学习偏移量和聚类带宽的同时细胞分割和跟踪

作者：Katharina Löffler,Ralf Mikut

机构*：!

链接：点击下载PDF文件

【4】 Dynamic Prototype Convolution Network for Few-Shot Semantic Segmentation

标题：基于动态原型卷积网络的Few-Shot语义分割

作者：Jie Liu,Yanqi Bao,Guo-Sen Xie,Huan Xiong,Jan-Jakob Sonke,Efstratios Gavves

机构*：University of Amsterdam, Netherlands, Northeastern University, China, The Netherlands Cancer Institute, Netherlands, Nanjing University of Science and Technology, China, Mohamed bin Zayed University of Artificial Intelligence, UAE

备注：Accepted in CVPR2022. Code will be available soon

链接：点击下载PDF文件

【5】 Interactive Segmentation and Visualization for Tiny Objects in Multi-megapixel Images

标题：多百万像素图像中微小目标的交互式分割与可视化

作者：Chengyuan Xu,Boning Dong,Noah Stier,Curtis McCully,D. Andrew Howell,Pradeep Sen,Tobias Höllerer

机构*：Tobias H¨ollerer, University of California, Santa Barbara, Las Cumbres Observatory

备注：6 pages, 4 figures. Accepted by CVPR 2022 Demo Program

链接：点击下载PDF文件

【6】 Development of an algorithm for medical image segmentation of bone tissue in interaction with metallic implants

标题：与金属种植体相互作用的骨组织医学图像分割算法研究

作者：Fernando García-Torres,Carmen Mínguez-Porter,Julia Tomás-Chenoll,Sofía Iranzo-Egea,Juan-Manuel Belda-Lois

机构*：org 1 Instituto de Biomec´anica de Valencia (IBV) 2 Universitat Politecnica de Valencia (UPV)Carmen M´ınguez Porter 2carminpo

备注：3 pages, 2 figures, MIDL

链接：点击下载PDF文件

【Zero/Few Shot|迁移|域适配|自适应】(3篇)

【1】 Leveraging Deepfakes to Close the Domain Gap between Real and Synthetic Images in Facial Capture Pipelines

标题：利用Deepfakes缩小面部捕捉管道中真实图像和合成图像之间的域差距

作者：Winnie Lin,Yilin Zhu,Demi Guo,Ron Fedkiw

机构*：Stanford University, Epic Games

链接：点击下载PDF文件

【2】 Multimodal Adaptive Distillation for Leveraging Unimodal Encoders for Vision-Language Tasks

标题：用于视觉语言任务的利用单峰编码器的多模式自适应蒸馏

作者：Zhecan Wang,Noel Codella,Yen-Chun Chen,Luowei Zhou,Xiyang Dai,Bin Xiao,Jianwei Yang,Haoxuan You,Kai-Wei Chang,Shih-fu Chang,Lu Yuan

机构*： Columbia University, Microsoft Research, University of California, Los Angeles

备注：arXiv admin note: substantial text overlap with arXiv:2201.05729

链接：点击下载PDF文件

【3】 Contrastive Test-Time Adaptation

标题：对比测试时间适应

作者：Dian Chen,Dequan Wang,Trevor Darrell,Sayna Ebrahimi

机构*：Toyota Research Institute, UC Berkeley

备注：CVPR 2022 camera-ready version

链接：点击下载PDF文件

【半弱无监督|主动学习|不确定性】(2篇)

【1】 DiRA: Discriminative, Restorative, and Adversarial Learning for Self-supervised Medical Image Analysis

标题：DIRA：用于自我监督医学图像分析的辨别性、恢复性和对抗性学习

作者：Fatemeh Haghighi,Mohammad Reza Hosseinzadeh Taher,Michael B. Gotway,Jianming Liang

机构*：Arizona State University

备注：Accepted at CVPR 2022 [main conference]

链接：点击下载PDF文件

【2】 PreTraM: Self-Supervised Pre-training via Connecting Trajectory and Map

标题：PreTraM：通过连接轨迹和地图进行自我监督的预训练

作者：Chenfeng Xu,Tian Li,Chen Tang,Lingfeng Sun,Kurt Keutzer,Masayoshi Tomizuka,Alireza Fathi,Wei Zhan

机构*： University of California, Berkeley, Google Research, University of California San Diego

备注：The code is available at this https URL

链接：点击下载PDF文件

【时序|行为识别|姿态|视频|运动估计】(3篇)

【1】 Gen6D: Generalizable Model-Free 6-DoF Object Pose Estimation from RGB Images

标题：Gen6D：基于RGB图像的可推广的无模型六自由度物体姿态估计

作者：Yuan Liu,Yilin Wen,Sida Peng,Cheng Lin,Xiaoxiao Long,Taku Komura,Wenping Wang

机构*： The University of Hong Kong, Zhejiang University, Tencent, Texas A&M University

备注：Project page: this https URL

链接：点击下载PDF文件

【2】 Dite-HRNet: Dynamic Lightweight High-Resolution Network for Human Pose Estimation

标题：Dite-HRNet：动态轻量级高分辨率人体姿态估计网络

作者：Qun Li,Ziyi Zhang,Fu Xiao,Feng Zhang,Bir Bhanu

机构*：School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, China, Department of Electrical and Computer Engineering, University of California at Riverside, CA, USA

备注：Accepted by IJCAI-ECAI 2022

链接：点击下载PDF文件

【3】 SE-GAN: Skeleton Enhanced GAN-based Model for Brush Handwriting Font Generation

标题：SE-GAN：基于骨架增强型GAN的毛笔字库生成模型

作者：Shaozu Yuan,Ruixue Liu,Meng Chen,Baoyang Chen,Zhijie Qiu,Xiaodong He

机构*：JD AI, Beijing, China ,Central Academy of Fine Arts, China

备注：Accepted by ICME 2022

链接：点击下载PDF文件

【医学相关】(3篇)

【1】 Denoising of Three-Dimensional Fast Spin Echo Magnetic Resonance Images of Knee Joints using Spatial-Variant Noise-Relevant Residual Learning of Convolution Neural Network

标题：基于卷积神经网络空间变异噪声相关残差学习的膝关节三维快速自旋回波磁共振图像去噪

作者：Shutian Zhao,Donal G. Cahill,Siyue Li,Fan Xiao,Thierry Blu,James F Griffith,Weitian Chen

机构*：. Department of Imaging and Interventional Radiology, the Chinese University of Hong, . Department of Electrical Engineering, the Chinese University of Hong Kong, Running title: Residual Learning of Neural Network Denoising for Dual-NEX Three-

备注：6 figures, abstract accepted by Joint Annual Meeting ISMRM-ESMRMB & ISMRT 31st Annual Meeting

链接：点击下载PDF文件

【2】 MIPR:Automatic Annotation of Medical Images with Pixel Rearrangement

标题：MIPR：基于像素重排的医学图像自动标注

作者：Pingping Dai,Haiming Zhu,Shuang Ge,Ruihan Zhang,Xiang Qian,Xi Li,Kehong Yuan

机构*：Tsinghua University ,Peking University Shenzhen Hospital

链接：点击下载PDF文件

【3】 Scale-Equivariant Unrolled Neural Networks for Data-Efficient Accelerated MRI Reconstruction

标题：基于尺度等变展开神经网络的数据高效加速MRI重建

作者：Beliz Gunel,Arda Sahiner,Arjun D. Desai,Akshay S. Chaudhari,Shreyas Vasanawala,Mert Pilanci,John Pauly

机构*： Stanford University, CA, USA, Equal Contribution

链接：点击下载PDF文件

【GAN|对抗|攻击|生成相关】(3篇)

【1】 Centralized Adversarial Learning for Robust Deep Hashing

标题：基于集中式对抗性学习的健壮深度散列算法

作者：Xunguang Wang,Xu Yuan,Zheng Zhang,Guangming Lu,Xiaomeng Li

机构*：Li, Harbin Institute of Technology, Shenzhen

链接：点击下载PDF文件

【2】 Enhancing the Transferability via Feature-Momentum Adversarial Attack

标题：通过特征-动量对抗性攻击增强可转移性

作者：Xianglong,Yuezun Li,Haipeng Qu,Junyu Dong

机构*：Ocean University of China

链接：点击下载PDF文件

【3】 Recurrent Affine Transformation for Text-to-image Synthesis

标题：用于文本到图像合成的递归仿射变换

作者：Senmao Ye,Fei Liu,Minkui Tan

机构*：South China University of Technology

链接：点击下载PDF文件

【Attention注意力】(1篇)

【1】 Attentions Help CNNs See Better: Attention-based Hybrid Image Quality Assessment Network

标题：关注度帮助CNN看得更清楚：基于关注度的混合图像质量评估网络

作者：Shanshan Lao,Yuan Gong,Shuwei Shi,Sidi Yang,Tianhe Wu,Jiahao Wang,Weihao Xia,Yujiu Yang

机构*： Tsinghua Shenzhen International Graduate School, Tsinghua University, University College London

链接：点击下载PDF文件

【人脸|人群计数】(1篇)

【1】 JIFF: Jointly-aligned Implicit Face Function for High Quality Single View Clothed Human Reconstruction

标题：Jiff：用于高质量单视人体重建的联合对齐隐式人脸函数

作者：Yukang Cao,Guanying Chen,Kai Han,Wenqi Yang,Kwan-Yee K. Wong

机构*：The University of Hong Kong, The Future Network of Intelligence Institute (FNii), CUHK-Shenzhen

备注：Camera-ready for CVPR 2022. Project page: this https URL

链接：点击下载PDF文件

【跟踪】(1篇)

【1】 Improving tracking with a tracklet associator

标题：使用tracklet关联器改进跟踪

作者：Rémi Nahon,Guillaume-Alexandre Bilodeau,Gilles Pesant

机构*：Department of Computer Engineering and Software Engineering, Polytechnique Montréal, Montréal, Canada

备注：8 pages, 6 figures, CRV 2022

链接：点击下载PDF文件

【裁剪|量化|加速|压缩相关】(1篇)

【1】 Depth Pruning with Auxiliary Networks for TinyML

标题：基于辅助网络的TinyML深度剪枝

作者：Josen Daniel De Leon,Rowel Atienza

机构*：⋆Electrical and Electronics Engineering Institute, University of the Philippines, †Samsung Research Philippines

备注：To be published in International Conference on Acoustics, Speech, & Signal Processing (ICASSP) 2022

链接：点击下载PDF文件

【视觉解释|视频理解VQA|caption等】(1篇)

【1】 Learning to Scaffold: Optimizing Model Explanations for Teaching

标题：学习脚手架：优化教学模式讲解

作者：Patrick Fernandes,Marcos Treviso,Danish Pruthi,André F. T. Martins,Graham Neubig

机构*：ΨLanguage Technologies Institute, Carnegie Mellon University, Pittsburgh, PA, ΩInstituto Superior Técnico & LUMLIS (Lisbon ELLIS Unit), Lisbon, Portugal, ℜInstituto de Telecomunicações, Lisbon, Portugal, ΛAmazon Web Services, ΓUnbabel, Lisbon, Portugal

备注：9 pages

链接：点击下载PDF文件

【点云|SLAM|雷达|激光|深度RGBD相关】(4篇)

【1】 PU-EVA: An Edge Vector based Approximation Solution for Flexible-scale Point Cloud Upsampling

标题：PU-EVA：一种基于边缘向量的柔性点云上采样近似解

作者：Luqing Luo,Lulu Tang,Wanyi Zhou,Shizheng Wang,Zhi-Xin Yang

机构*：State Key Laboratory of Internet of Things for Smart City, University of Macau, South China University of Technology, Institute of Microelectronics Chinese Academy of Sciences

链接：点击下载PDF文件

【2】 Reconstructing Surfaces for Sparse Point Clouds with On-Surface Priors

标题：基于面上先验的稀疏点云曲面重建

作者：Baorui Ma,Yu-Shen Liu,Zhizhong Han

机构*：School of Software, BNRist, Tsinghua University, Beijing, China, Wayne State University, Detroit, USA

备注：To appear at CVPR2022. Project page:this https URL this https URL

链接：点击下载PDF文件

【3】 Learning Dynamic View Synthesis With Few RGBD Cameras

标题：用较少的RGBD摄像机学习动态视图合成

作者：Shengze Wang,YoungJoong Kwon,Yuan Shen,Qian Zhang,Andrei State,Jia-Bin Huang,Henry Fuchs

机构*： UNC-Chapel Hill, UIUC, Virginia Tech

链接：点击下载PDF文件

【4】 Monocular Depth Estimation Using Cues Inspired by Biological Vision Systems

标题：基于生物视觉系统启发的单眼深度估计

作者：Dylan Auty,Krystian Mikolajczyk

机构*：MatchLab Group, Electrical and Electronic Engineering Department, Imperial College London

备注：7 pages, 2 figures. Accepted to International Conference on Pattern Recognition (ICPR) 2022. Code available at this https URL

链接：点击下载PDF文件

【多模态】(1篇)

【1】 Learning Sequential Latent Variable Models from Multimodal Time Series Data

标题：从多峰时间序列数据中学习序贯潜变量模型

作者：Oliver Limoyo,Trevor Ablett,Jonathan Kelly

机构*： University of Toronto Institute for Aerospace Studies (UTIAS), Toronto, Canada, ∗Corresponding author

备注：Accepted to the International Conference on Intelligent Autonomous Systems (IAS17), Zagreb, Croatia, June 13-16, 2022

链接：点击下载PDF文件

【其他神经网络|深度学习|模型|建模】(3篇)

【1】 Exposure Correction Model to Enhance Image Quality

标题：一种提高图像质量的曝光校正模型

作者：Fevziye Irem Eyiokur,Dogucan Yaman,Hazım Kemal Ekenel,Alexander Waibel

机构*：Karlsruhe Institute of Technology,Istanbul Technical University,Carnegie Mellon University

备注：Accepted for CVPR2022 NTIRE Workshop

链接：点击下载PDF文件

【2】 A Note on the Regularity of Images Generated by Convolutional Neural Networks

标题：关于卷积神经网络生成图像正则性的一个注记

作者：Andreas Habring,Martin Holler

机构*：Institute of Mathematics and, Scientific Computing, University of Graz, Graz, Austria

链接：点击下载PDF文件

【3】 Fourier Imager Network (FIN): A deep neural network for hologram reconstruction with superior external generalization

标题：傅里叶成像网络(FIN)：一种用于全息重建的深度神经网络，具有优越的外部泛化能力

作者：Hanlong Chen,Luzhe Huang,Tairan Liu,Aydogan Ozcan

机构*： Electrical and Computer Engineering Department, University of California, Los Angeles, California, Bioengineering Department, University of California, Los Angeles, California , USA

备注：19 Pages, 5 Figures, 1 Table

链接：点击下载PDF文件

【其他】(8篇)

【1】 Control-NeRF: Editable Feature Volumes for Scene Rendering and Manipulation

标题：Control-Nerf：场景渲染和操纵的可编辑特征体积

作者：Verica Lazova,Vladimir Guzov,Kyle Olszewski,Sergey Tulyakov,Gerard Pons-Moll

机构*：University of T¨ubingen, Max Planck Institute for Informatics, Saarland Informatics Campus, Snap Inc., b) Inserting objects from one scene into another, T-rex inserted inside the garden scene, Second T-rex added to the scene, a) Original scenes

链接：点击下载PDF文件

【2】 SUES-200: A Multi-height Multi-scene Cross-view Image Benchmark Across Drone and Satellite

标题：SUES-200：一种跨越无人机和卫星的多高度多场景交叉图像基准

作者：Runzhe Zhu

机构*：Runzhe Zhu is with the School of Electronic and Electrical Engineer, Shanghai University of Engineering Science

链接：点击下载PDF文件

【3】 Balancing Expert Utilization in Mixture-of-Experts Layers Embedded in CNNs

标题：在CNN中嵌入的专家混合层中平衡专家利用率

作者：Svetlana Pavlitskaya,Christian Hubschneider,Lukas Struppek,J. Marius Zöllner

机构*：FZI Research Center for Information Technology, Karlsruhe, Germany, Department of Computer Science, Technical University of Darmstadt, Darmstadt, Germany

链接：点击下载PDF文件

【4】 Spacing Loss for Discovering Novel Categories

标题：发现新类别的间距损失

作者：K J Joseph,Sujoy Paul,Gaurav Aggarwal,Soma Biswas,Piyush Rai,Kai Han,Vineeth N Balasubramanian

机构*：Indian Institute of Technology Hyderabad, Google Research, Indian Institute of Science, Indian Institute of Technology Kanpur, The University of Hong Kong

备注：Accepted to Continual Learning in Computer Vision Workshop (CLVision) at CVPR 2022

链接：点击下载PDF文件

【5】 Data Clustering as an Emergent Consensus of Autonomous Agents

标题：数据聚类作为自治智能体的一种紧急共识

作者：Piotr Minakowski,Jan Peszek

机构*：Institute of Analysis and Numerics, Otto von Guericke University Magdeburg, Universit¨atsplatz , Magdeburg, Germany, Institute of Applied Mathematics and Mechanics, University of Warsaw, ul. Banacha ,-, Warszawa, Poland

链接：点击下载PDF文件

【6】 Alleviating Representational Shift for Continual Fine-tuning

标题：缓解表征漂移以实现连续微调

作者：Shibo Jie,Zhi-Hong Deng,Ziheng Li

机构*：School of Artificial Intelligence, Peking University

链接：点击下载PDF文件

【7】 Keypoint based Sign Language Translation without Glosses

标题：基于关键点的无注释手语翻译

作者：Youngmin Kim,Minji Kwak,Dain Lee,Yeongeun Kim,Hyeongboo Baek

机构*：Incheon National University, Incheon, Republic of Korea, Chung-Ang University, Seoul, Republic of Korea, Ewha Womans University, Konkuk University

备注：14 pages, 5 figures, IEEE Sensors Journals

链接：点击下载PDF文件

【8】 The 6th AI City Challenge

标题：第六届AI城市挑战赛

作者：Milind Naphade,Shuo Wang,David C. Anastasiu,Zheng Tang,Ming-Ching Chang,Yue Yao,Liang Zheng,Mohammed Shaiqur Rahman,Archana Venkatachalapathy,Anuj Sharma,Qi Feng,Vitaly Ablavsky,Stan Sclaroff,Pranamesh Chakraborty,Alice Li,Shangru Li,Rama Chellappa

机构*： NVIDIA Corporation, CA, USA, Santa Clara University, CA, USA, University at Albany, SUNY, NY, USA, Australian National University, Australia, Indian Institute of Technology Kanpur, India, Iowa State University, IA, USA, Boston University, MA, USA

备注：Summary of the 6th AI City Challenge Workshop in conjunction with CVPR 2022

链接：点击下载PDF文件

机器翻译，仅供参考

重大更新！公众号每日速递覆盖arXiv所有方向,涵盖CS|物理|数学|经济|统计|金融|生物|电气等领域。点击查看历史文章列表，可在公众号内订阅#arXiv每日学术速递话题，接收每日推送。点击访问arxivdaily.com每日更新更及时。