使用基於 Tensorflow 下的預訓練模型「COCOSSD」當作物體辨識的基礎,該模型 可提供種類豐富的資料集,利用電腦的 webcam 當作攝影機,並透過攥寫 ml5.js,即能 使用瀏覽器當作媒介將辨識結果逐幀顯示出來。
The pre training model "COCOSSD" based on Tensorflow is used as the basis of object recognition. This model can provide a variety of data sets. The computer webcam is used as the camera, and by writing ml5.js, the browser can be used as the medium to display the recognition results frame by frame. (2024-03-21, HTML, 0KB, 下载0次)
自動車のナンハ?ーフ?レートを読み込むサンフ?ルフ?ロク?ラムて?す。USBカメラて?ナンハ?ーフ?レートを読み込み、その結果をテキストテ?ータとして扱うことか?て?きます。,
The automatic train is in the form of a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, a car, and a car. USB 扱ことか???きす., (2022-04-16, C#, 0KB, 下载0次)
本專題主要使用 OpenCV 透過自行訓練並建立 Haar 特徵分類器「偵測車牌」模型,先偵測框選車牌及原始圖片轉換尺寸及偵測、然後將擷取車牌號碼產生圖形、再來是去除畸零地把邊緣的輪廓、雜訊、黑色部分去除、最後將用光學辨識軟體 OCR 辨識出車牌號碼,就會出現以下結果及結論。,
This topic mainly uses OpenCV to train itself and establish the Haar feature classifier "Detect License Plate" model. First, detect the frame selected license plate and the original image conversion size and detection, then extract the license plate number to generate graphics, and then remove the edge contour, noise, black parts, and finally identify the license plate number with optical identification software OCR. The following results and conclusions will appear., (2023-08-23, Jupyter Notebook, 0KB, 下载0次)
这是达里奥·索蒂诺（Dario Sortino）的工作，安德烈亚·博纳里尼（Andrea Bonarini）教授建议他检测孩子的姿势和注视方向...
This is the work done by Dario Sortino, advised by prof.Andrea Bonarini to detect pose of a kid and gaze direction towards a recognized object to implement shared attention mechanisms with children with ASD. The DeepNN are implemented on Google Coral and Intel MovidiusII (2021-09-28, C++, 48559KB, 下载0次)
伊万·里卡特·博尔赫斯（Ivan Ricart Borges）-使用Suprema API通过crea...
Ivan Ricart Borges - Using the Suprema API to synchronize all workers fingerprints with all biometric devices by creating .NET components to interact with the web platform. Use ActiveX technology to incorporate fingerprint register functionality in web environment. (2021-05-15, C#, 3839KB, 下载0次)
在各种不同的语言中提供的免费编程书籍无疑是GitHub上最受欢迎和最受欢迎的资料库之一。尽管它的名字里有“书”,但它提供的远不止这些。它包含了免费在线课程、互动编程资源、问题集和竞赛性编程、播客和几乎每种编程语言的学习列表。
The free programming books available in various languages are undoubtedly one of the most popular and popular databases on GitHub. Although its name includes book , it provides far more than that. It includes free online courses, interactive programming resources, problem sets and competitive programming, podcasts, and learning lists for almost every programming language. (2022-05-24, Others, 485KB, 下载0次)
2020.3.9 创建的iOS项目,主要用于研究当下热门的技术点功能,使用第三方sdk实现 (环信)自定义UI的即时通讯(防微信),(即构,声网)自定义UI的直播功能(映客的真人秀直播,QQMusic的语音直播),(美摄)自定义UI的美颜...
The iOS project created on March 9, 2022 is mainly used to research current popular technology features, using third-party SDKs to implement instant messaging (WeChat prevention) for custom UI, live streaming functions for custom UI (Yingke s reality show live streaming, QQMusic s voice live streaming), and beauty for custom UI (2020-03-16, Objective-C, 16KB, 下载0次)
Skin-Clothes-Hair-Segmentation-using-SMP,3クラス(肌、服、髪)のセマンティックセク?メンテーションを実施するモテ?ル(A model that performs semantic segmentation of 3 classes(skin, clothes, hair))
Skin Lothes Hair Segmentation Using SMP, 3 クラススセマンィックセク?メン?ションをソるモ A model that performs semantic segmentation of 3 classes (skin, clothes, hair) (2021-12-20, Jupyter Notebook, 79845KB, 下载0次)
ddlee-tech-weekly,汇总每周我认为值得分享的信息,以“原链接+评论”的方式呈现,分为博文、开源项目、技术简讯等版块,关注的主要方向为AI、计算机视觉、Android、Linux及泛互联网话题。
Middle tech weekly, summarizing information that I think is worth sharing every week, presented in the form of "original links+comments", divided into sections such as blogs, open source projects, technical newsletters, etc. The main focus is on AI, computer vision, Android, Linux, and pan internet topics. (2019-06-10, HTML, 149KB, 下载0次)
Art_Blog,基于Spring+SpringMVC+Mybatis架构的开源博客:博客管理、图表数据、日志分析、访问记录、图库管理、资源管理、友链通知等。良好的页面预加载,无限滚动加载,文章置顶,博主推荐等。提供 用户端+管理端 的整套系统源码。响...
Art_ Blog, an open-source blog based on the Spring+SpringMVC+Mybatis architecture: blog management, chart data, log analysis, access records, graph library management, resource management, friend chain notifications, etc. Good page preloading, unlimited scrolling loading, article topping, blogger recommendations, etc. Provide a complete set of system source code for both user and management ends. Loud (2018-12-14, JavaScript, 9307KB, 下载0次)
酒店管理系统，。。。服务。你可以添加用户和客户，删除他们，并根据他们的姓名和电话号码搜索用户。#在...中运行pip...
The Hotel Management System Project is a general software developed to simplify hotel operations by automating them. the project has Login System security with database in MySql. This project collects information about customers, room and other hotel services. you can add user and costomers and delete them and search user as per there name (2022-04-30, Python, 1646KB, 下载0次)
cntk酒店图片分类器，该POC使用cntk 2.1来训练图像多类别分类的模型。我们的模型能够识别特殊...
This POC is using CNTK 2.1 to train model for multiclass classification of images. Our model is able to recognize specific objects (i.e. toilet, tap, sink, bed, lamp, pillow) connected with picture types we are looking for. It plays a big role in a process which will be used to classify pictures from different hotels and determine whether it s (2018-02-15, Python, 2452KB, 下载0次)
某汽车租赁公司需要一个租车管理系统。要求具有以下功能：
<1. 管理车辆基本信息，增加/删除/列表；
<2. 能进行租车/还车操作；
<3. 按车辆类别统计租金(虽然是选做，已做好)。
车辆基本信息包括：车辆类别（大客/面包/轿车），品牌型号，车牌号，租车价格
租车时应记录：租车日期，车辆id，客户名称；
还车时应记录：还车日期。
A need for a car rental company car rental management system. Required to have the following features:
<A basic vehicle information management, add/delete/list
<2 can be car/truck operators also
<3 by vehicle category statistics rent (although elected to do, have to do).
Basic vehicle information includes: Vehicle category (large passenger/bread/sedan), brand model, license plate number, car rental prices
Car should record: Car date, vehicle id, customer name
Also the car should be recorded: the car date. (2014-04-02, Java, 3120KB, 下载13次)
包括中西方人的人脸性别识别。基于局部Gabor变换直方图序列(LGBPHS)的方法。图像噪声、光照和人脸角度变化对性别识别的影响都非常小。
尤其对东方人的性别识别率可达96
Including Westerners who face gender identification. Based on local Gabor transform histogram sequence (LGBPHS), method. Image noise, light and face angle change gender identification are very small. Especially Asians gender recognition rate up to 96 (2013-03-21, matlab, 26032KB, 下载321次)
这篇文章主要是介绍多个手部的分割，是在前面的博文：不需要骨骼跟踪的人体手部分割 的基础上稍加改进的。因为识别有的一个应用场合就是手势语言识别，而手势一般都需要人的2只手相配合完成，因此很有必要对人体的多个手部来进行分割。
The article introduced hand split, in front of the Bowen: does not require tracking of human hand bones split on the basis of a little improvement. Because the identification of an applications is the sign language recognition, gesture generally requires two people hand cooperates completed, so it is necessary to perform the segmentation of the body of a plurality of hand. (2012-11-01, C/C++, 1113KB, 下载95次)
手势识别非常重要的一个特点是要体验要好，即需要以用户为核心。而手势的定位一般在手势识别过程的前面，在上一篇博文Kinect+OpenNI学习笔记之8(Robert Walter手部提取代码的分析) 中已经介绍过怎样获取手势区域，且取得了不错的效果，但是那个手势部位的提取有一个大的缺点，即需要人站立起来，当站立起来后才能够分隔出手。而手势在人之间的交流时，并不一定要处于站立状态，所以这不是一个好的HCI。因此本文介绍的手势部位的提取并不需要人处于站立状态，同样取得了不错的效果。
A characteristic gesture recognition is very important to experience better, and that needs to be user core. Gestures positioning generally in front of the gesture recognition process on a blog Kinect+OpenNI learning notes (Robert Walter hands to extract the code analysis) has been described how to obtain the gesture area, and achieved good results, but that gestures parts extraction have a big drawback, namely the need to stand up, be able to stand up before separated shot. While the gesture when the exchanges between the people, not necessarily to be in a standing position, so this is not a good HCI. So part of the extract of the article describes the gesture does not need people in the standing position, also achieved good results. (2012-10-19, C/C++, 900KB, 下载224次)