[go: up one dir, main page]

CN1456970A - How to replace objects in dynamic images - Google Patents

How to replace objects in dynamic images Download PDF

Info

Publication number
CN1456970A
CN1456970A CN 02119710 CN02119710A CN1456970A CN 1456970 A CN1456970 A CN 1456970A CN 02119710 CN02119710 CN 02119710 CN 02119710 A CN02119710 A CN 02119710A CN 1456970 A CN1456970 A CN 1456970A
Authority
CN
China
Prior art keywords
dynamic image
objects
database
image
replacing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN 02119710
Other languages
Chinese (zh)
Inventor
邓恒敦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Aiptek International Inc
Original Assignee
Aiptek International Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Aiptek International Inc filed Critical Aiptek International Inc
Priority to CN 02119710 priority Critical patent/CN1456970A/en
Publication of CN1456970A publication Critical patent/CN1456970A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Processing Or Creating Images (AREA)

Abstract

The invention relates to a method for replacing objects in a dynamic image, wherein the dynamic image is a dynamic image file or a real-time dynamic image, a program is used for replacing the objects of the dynamic image, the method comprises a first database, a second database and a user interface, the content stored in the first database is a plurality of characteristics of a first object to be replaced in the dynamic image, the content stored in the second database is a second object to be replaced in the first object, when the dynamic image is played, the object characteristics in the first database start to search the first object in the dynamic image and carry out a program of object characteristic comparison, the first object contained in the dynamic image is searched and judged by a similarity, when the similarity is more than a set value, the searched object is judged to be the first object, then the second object stored in the second database is used for replacing the first object in the dynamic image being played, during the playing process, the user interface can be used to capture a static image of the first object replaced by the second object.

Description

动态影像内的物件的替换方法How to replace objects in dynamic images

技术领域technical field

本发明涉及一种影像内的物件的替换方法,特别是一种动态影像内的物件的替换方法,其中此动态影像为一动态影像档或是即时动态影像,以提供使用者有更多的选择以替代即时动态影像与动态影像内的物件,达到更佳的娱乐效果。The present invention relates to a method for replacing objects in an image, especially a method for replacing objects in a dynamic image, wherein the dynamic image is a dynamic image file or a real-time dynamic image to provide users with more choices To replace real-time dynamic images and objects in the dynamic images to achieve better entertainment effects.

背景技术Background technique

对现今的信息使用者,尤其是个人电脑使用者而言,所接触电脑信息的模式已由较为单调的文字模式而进入到所谓的多媒体(Multi-Media)的时代,而多媒体意味着资料包含文字、影像、声音等内容,而为了使资料能以多媒体的形式呈现在观众的面前,现今已发展出多种可撷取影像或是声音等的多媒体装置。因为多媒体资料包含较多的内容,因此处理多媒体资料所需要的记忆体的容量需较处理单调的文字模式为高,以使多媒体资料以更为顺畅的模式呈现出来。For today's information users, especially personal computer users, the mode of contacting computer information has entered the era of so-called Multi-Media from the relatively monotonous text mode, and multimedia means that the information includes text , video, audio and other content, and in order to present the data in the form of multimedia in front of the audience, a variety of multimedia devices that can capture video or audio have been developed. Because the multimedia data contains more content, the memory capacity required to process the multimedia data needs to be higher than the monotonous text mode, so that the multimedia data can be presented in a smoother mode.

对影像输入装置而言,扫描器(scanner)、数字摄影机(digital videocamera)、数字照相机(digital camera)为目前较为常见的装置。数字照相机是利用数字化的信息来储存所拍摄得的静态影像。而数字摄影机则是利用数字化的信息来储存所拍摄得的静态影像。静态影像及动态影像的数据皆利用一个可记录影像的磁盘或是个人电脑记忆体卡国际协会(Personal Computer Memory Card International Association;PCMCIA)的记忆卡,来储存影像数据。为了节省储存媒体所使用的记忆空间,通常数字照相机所拍摄的影像都是以压缩档案来储存,常见的有静态影像压缩标准(Joint Photographic Experts Group;JPEG)档案格式。而数字摄影机则通常采用AVI多媒体播放格式来储存。For image input devices, scanners, digital video cameras, and digital cameras are relatively common devices at present. Digital cameras use digitized information to store captured still images. The digital video camera uses digital information to store the captured still images. The data of the still image and the dynamic image all use a disk that can record the image or a memory card of the Personal Computer Memory Card International Association (PCMCIA) to store the image data. In order to save the memory space used by the storage media, images captured by digital cameras are usually stored in compressed files, and the common one is the still image compression standard (Joint Photographic Experts Group; JPEG) file format. Digital video cameras usually use the AVI multimedia playback format to store.

AVI多媒体播放格式是微软公司(Microsoft)在针对视窗系统所开发出来的影音档(Video for Windows)当中所定义的档案格式。属于AVI多媒体播放格式的档案当中,可存放许多不同种类媒体的资料流(data stream)。每个资料流都可以是一种独立的媒体,例如:影像资料流、声音资料流、文字资料流等。而支援AVI多媒体播放格式的应用程序,必须能将这些资料流做同步播放的工作。因此AVI多媒体播放格式是在多媒体领域当中最重要的档案格式之一。The AVI multimedia playback format is a file format defined by Microsoft Corporation (Microsoft) in the video and audio files (Video for Windows) developed for the window system. Among the files belonging to the AVI multimedia playback format, data streams (data streams) of many different types of media can be stored. Each data stream can be an independent medium, for example: image data stream, audio data stream, text data stream, etc. The application programs that support the AVI multimedia playback format must be able to perform synchronous playback of these data streams. Therefore, the AVI multimedia playback format is one of the most important file formats in the multimedia field.

AVI档案不须特别的硬件设备便可播放,因此在个人电脑(personal computer;PC)上流传得十分广泛。因为此种档案格式是将动画分割成为一张张静态影像,后面的影像只储存和前面影像差异的部分,因此在播放之前必须要做某些图素比较的动作,每秒约只能播放20个画面影像,而且这些影像通常只有半个屏幕大小。AVI files can be played without special hardware equipment, so they are widely circulated on personal computers (personal computers; PCs). Because this file format divides the animation into static images, and only the difference between the subsequent images and the previous images is stored, so some pixel comparisons must be done before playback, and only about 20 images can be played per second. frame images, and these images are usually only half the size of the screen.

一般数字摄影机内中的档案传递至电脑后可储存为一AVI多媒体格式的档案。而将一电荷耦合元件(Charge-Coupled Device;CCD)连接至一电脑上时,可在屏幕上得到一即时动态影像。所谓的电荷耦合元件是一种对光线敏感的电路装置。数字照相机或是数字摄影机便是利用残留在电荷耦合元件上面的讯息,将每个图点元件对光的敏感度转换为数字影像资料而储存。The files in the general digital video camera can be stored as a file in AVI multimedia format after being transferred to the computer. And when a Charge-Coupled Device (CCD) is connected to a computer, a real-time dynamic image can be obtained on the screen. A so-called charge-coupled device is a light-sensitive circuit arrangement. A digital camera or a digital video camera uses the information remaining on the charge-coupled device to convert the sensitivity of each dot element to light into digital image data for storage.

在传统的方法中,当播放一多媒体格式的动态影像档案时,若欲隐藏此动态影像档案内部分的物件,诸如:人脸,则通常采取一雾状的方式,将此雾状的格式与动态影像档案内的部分元件相互重叠,以使观赏者无法清楚地看出部分元件的真实面貌。但是此雾状的方法无法使用在即时动态影像档案上,且此雾状方法会因为雾状程度的设定值不同而无法充分保护欲隐藏的物件的全貌。而且此雾状的方法将可由软件或是硬件轻易地破解而失去了隐藏部分物件的意义。因此传统采用雾状的方式以隐蔽动态影像档案内部分的物件将有可能造成使用者的不便,且无法达到娱乐大众的效果。In the traditional method, when playing a dynamic image file in a multimedia format, if you want to hide some objects in the dynamic image file, such as a human face, you usually use a foggy method to combine the foggy format with the Some elements in the dynamic image file overlap each other, so that the viewer cannot clearly see the real appearance of some elements. However, this fogging method cannot be applied to real-time dynamic image files, and this fogging method cannot fully protect the full picture of the object to be hidden due to the different setting values of the fogging degree. Moreover, this foggy method can be easily cracked by software or hardware and loses the meaning of hiding some objects. Therefore, the traditional method of using fog to conceal some objects in the dynamic image file may cause inconvenience to the user and fail to achieve the effect of entertaining the public.

发明内容Contents of the invention

本发明的目的在于克服现有技术中的不足与缺陷,即传统的雾状处理法无法确实隐蔽即时动态影像内的部分物件,从而提供了一种动态影像内的物件的替换方法,利用资料库中特征的辨识与物件的取代替换动态影像档内的部分物件,以确实隐蔽动态影像内所欲隐蔽的部分物件,其中此动态影像为一动态影像档或是即时动态影像。The purpose of the present invention is to overcome the deficiencies and defects in the prior art, that is, the traditional fog processing method cannot really conceal some objects in the real-time dynamic image, thereby providing a method for replacing objects in the dynamic image, using the database Recognition of features and replacement of objects to replace some objects in the dynamic image file, so as to truly conceal the desired hidden part of the object in the dynamic image, wherein the dynamic image is a dynamic image file or a real-time dynamic image.

本发明的第二个目的为利用资料库中特征的辨识与物件的取代替换即时动态影像档案与动态影像档案内的部分物件,以增加动态影像的娱乐效果,其中此动态影像为一动态影像档或是即时动态影像。The second purpose of the present invention is to use the identification of features in the database and the replacement of objects to replace real-time dynamic image files and some objects in the dynamic image files, so as to increase the entertainment effect of dynamic images, wherein the dynamic image is a dynamic image file Or live video.

本发明的第三个目的为利用资料库中特征的辨识与物件的取代替换动态影像内的部分物件,以充分保护个人的隐私,其中此动态影像为一动态影像档或是即时动态影像。The third purpose of the present invention is to fully protect the privacy of individuals by using the identification of features in the database and the replacement of objects to replace some objects in the dynamic image, wherein the dynamic image is a dynamic image file or a real-time dynamic image.

本发明的再一个目的为利用资料库中特征的辨识与物件的取代替换动态影像内的部分物件,以加速影像处理的效果,其中此动态影像为一动态影像档或是即时动态影像。Another object of the present invention is to use the identification of features in the database and the replacement of objects to replace some objects in the dynamic image to accelerate the effect of image processing, wherein the dynamic image is a dynamic image file or real-time dynamic image.

根据以上所述的目的,本发明提供了一种方法,利用资料库中特征的辨识与物件的取代替换动态影像内的部分物件,以确实隐蔽动态影像内所欲隐蔽的部分物件,其中此动态影像为一动态影像档或是即时动态影像。本发明的替换方法为利用一程序以取代动态影像内的物件,其中上述的程序包含一第一资料库、一第二资料库与一使用者介面(user interface)。第一资料库储存的内容为在动态影像内所欲取代的第一物件的多个特征,其中此动态影像为一动态影像档或是即时动态影像。第二资料库储存的内容为欲取代第一物件的第二物件。在动态影像播放时,本发明的程序即由第一资料库内的物件特征开始搜寻动态影像内的第一物件并进行一物件特征对比的程序。在物件特征对比的程序中,本发明的程序由一相似度以搜寻动态影像内所包含的第一物件。当所搜寻的物件经由特征对比的程序中判断其相似度大于一设定值时,即判断所搜寻的物件为第一物件。接下来利用第二资料库内所储存的第二物件以取代正在播放中的动态影像内的第一物件。在动态影像播放的过程中,使用者可利用使用者介面撷取一以第二物件取代第一物件的静态影像。当动态影像播放完毕后,使用者可在使用者介面上选择是否要以第二物件的动态影像取代第一物件的动态影像。当欲取代的第一物件为多个第一物件且不同的第一物件需以不同的第二物件取代时,则可利用不同的多个第一物件在动态影像内所移动的轨迹来分类,并将同一类的第一物件以相同的第二物件所取代。本发明的方法也可以增加动态影像的娱乐效果并充分保护个人的隐私。本发明的方法更可加速影像处理的效果。According to the above-mentioned purpose, the present invention provides a method, using the identification of features in the database and the replacement of objects to replace some objects in the dynamic image, so as to conceal the desired concealed part of the object in the dynamic image, wherein the dynamic The image is a dynamic image file or a real-time dynamic image. The replacement method of the present invention is to use a program to replace the objects in the dynamic image, wherein the above-mentioned program includes a first database, a second database and a user interface. The content stored in the first database is a plurality of characteristics of the first object to be replaced in the dynamic image, wherein the dynamic image is a dynamic image file or a real-time dynamic image. The content stored in the second database is the second object to replace the first object. When the dynamic image is played, the program of the present invention searches for the first object in the dynamic image from the object characteristics in the first database and performs a comparison of object characteristics. In the process of object feature comparison, the program of the present invention searches for the first object contained in the dynamic image by a similarity. When the similarity of the searched object is determined to be greater than a set value through the feature comparison program, it is determined that the searched object is the first object. Next, the second object stored in the second database is used to replace the first object in the dynamic image being played. In the process of playing the dynamic image, the user can use the user interface to capture a static image in which the first object is replaced by the second object. After the dynamic image is played, the user can choose whether to replace the dynamic image of the first object with the dynamic image of the second object on the user interface. When the first object to be replaced is a plurality of first objects and different first objects need to be replaced with different second objects, the moving trajectories of the different first objects in the dynamic image can be used to classify, and replace the first object of the same class with the same second object. The method of the invention can also increase the entertainment effect of dynamic images and fully protect personal privacy. The method of the present invention can further accelerate the effect of image processing.

附图说明Description of drawings

图1为本发明的第一流程图;Fig. 1 is the first flowchart of the present invention;

图2为本发明的第二流程图。Fig. 2 is a second flowchart of the present invention.

具体实施方式Detailed ways

下面结合附图和实施例详细说明本发明的具体实施方式。The specific implementation manner of the present invention will be described in detail below in conjunction with the accompanying drawings and examples.

本发明提供了一种方法,利用资料库中特征的辨识与物件的取代替换动态影像档案内的部分物件,以确实隐蔽动态影像内所欲隐蔽的部分物件,其中此动态影像为一动态影像档或是即时动态影像。本发明的替换方法为利用一程序以取代动态影像内的物件,其中上述的程序包含一第一资料库、一第二资料库与一使用者介面(userinterface)。第一资料库储存的内容为在动态影像内所欲取代的第一物件的多个特征。第二资料库储存的内容为欲取代第一物件的第二物件。The present invention provides a method, which utilizes the identification of features in the database and the replacement of objects to replace some objects in a dynamic image file, so as to surely conceal some objects to be hidden in the dynamic image, wherein the dynamic image is a dynamic image file Or live video. The replacement method of the present invention is to use a program to replace the objects in the dynamic image, wherein the above-mentioned program includes a first database, a second database and a user interface. The content stored in the first database is a plurality of characteristics of the first object to be replaced in the dynamic image. The content stored in the second database is the second object to replace the first object.

当使用者欲使用本发明的方法以取代动态影像内的部分物件时,则首先必须先安装本发明的方法所提供的程序并驱动此程序。接下来使用者可在使用者介面上设定所欲取代的动态影像内的第一物件,并在使用者介面上选择欲取代第一物件的第二物件。当第一物件为多个不同物件时,使用者也可在使用者介面上依序选取不同的多个第二物件,以对应不同的多个第一物件。When the user intends to use the method of the present invention to replace some objects in the dynamic image, the program provided by the method of the present invention must first be installed and driven. Next, the user can set the first object in the dynamic image to be replaced on the user interface, and select the second object to replace the first object on the user interface. When the first object is a plurality of different objects, the user can also sequentially select a plurality of different second objects on the user interface to correspond to a plurality of different first objects.

接下来利用一播放程序执行动态影像的档案,诸如:AVI多媒体格式的档案,或是利用电荷耦合元件以在屏幕上显示出一即时动态影像。当动态影像正在播放时,本发明的程序即由第一资料库内的物件特征开始搜寻动态影像内的第一物件。在物件特征对比的程序中,本发明的程序由一相似度以搜寻动态影像内所包含的第一物件并进行一物件特征对比的程序。当所搜寻的物件经由特征对比的程序中判断其相似度大于一设定值时,即判断所搜寻的物件为第一物件。接下来利用第二资料库内所储存的第二物件以取代正在播放中的动态影像内的第一物件。在动态影像播放的过程中,使用者可利用使用者介面撷取一以第二物件取代第一物件的静态影像。当动态影像播放完毕后,使用者可在使用者介面上选择是否要以第二物件的动态影像取代第一物件的动态影像。当欲取代的第一物件为多个第一物件且不同的第一物件需以不同的第二物件取代时,则可利用不同的多个第一物件在动态影像内所移动的轨迹来分类,并将同一类的第一物件以相同的第二物件所取代。Next, a playback program is used to execute the file of the moving image, such as a file in the AVI multimedia format, or a charge-coupled device is used to display a real-time moving image on the screen. When the dynamic image is playing, the program of the present invention starts to search for the first object in the dynamic image from the object characteristics in the first database. In the object feature comparison program, the program of the present invention uses a similarity to search for the first object included in the dynamic image and performs an object feature comparison process. When the similarity of the searched object is determined to be greater than a set value through the feature comparison program, it is determined that the searched object is the first object. Next, the second object stored in the second database is used to replace the first object in the dynamic image being played. In the process of playing the dynamic image, the user can use the user interface to capture a static image in which the first object is replaced by the second object. After the dynamic image is played, the user can choose whether to replace the dynamic image of the first object with the dynamic image of the second object on the user interface. When the first object to be replaced is a plurality of first objects and different first objects need to be replaced with different second objects, the moving trajectories of the different first objects in the dynamic image can be used to classify, and replace the first object of the same class with the same second object.

参照图1所示,此为本发明的第一流程图。首先使用者必须先安装及驱动本发明的方法所提供的程序100。接下来使用者可在使用者介面上设定所欲取代的即时动态影像档案或是动态影像档案内的第一物件,并在使用者介面上选择欲取代第一物件的第二物件110。当第一物件为多个不同物件时,使用者也可在使用者介面上依序选取不同的多个第二物件,以对应不同的多个第一物件。接下来利用一播放程序执行动态影像的档案,诸如:AVI多媒体格式的档案,或是利用电荷耦合元件以在屏幕上显示出一即时动态影像120。当动态影像或是即时动态影像正在播放时,本发明的程序即由第一资料库内的物件特征开始搜寻动态影像内的第一物件并进行一物件特征对比的程序130。在物件特征对比的程序中,本发明的程序由一相似度以搜寻动态影像内所包含的第一物件,并判断此相似度是否大于一设定值140。当所搜寻的物件经由特征对比的程序中判断其相似度大于一设定值时,即判断所搜寻的物件为第一物件150。当所搜寻的物件经由特征对比的程序中判断其相似度不大于一设定值时,则判断所搜寻到的元件并非为所欲取代的第一元件160。接下来利用第二资料库内所储存的第二物件以取代正在播放的动态影像内的第一物件170。Referring to Fig. 1, this is the first flow chart of the present invention. Firstly, the user must install and drive the program 100 provided by the method of the present invention. Next, the user can set the real-time dynamic image file to be replaced or the first object in the dynamic image file on the user interface, and select the second object 110 to replace the first object on the user interface. When the first object is a plurality of different objects, the user can also sequentially select a plurality of different second objects on the user interface to correspond to a plurality of different first objects. Next, a playback program is used to execute the dynamic image file, such as a file in AVI multimedia format, or a charge-coupled device is used to display a real-time dynamic image 120 on the screen. When the moving image or the real-time moving image is playing, the program 130 of the present invention searches for the first object in the moving image based on the object characteristics in the first database and compares the object characteristics. In the object feature comparison program, the program of the present invention searches for the first object included in the dynamic image according to a similarity degree, and judges whether the similarity degree is greater than a set value 140 . When the similarity of the searched object is determined to be greater than a set value through the feature comparison program, it is determined that the searched object is the first object 150 . When it is determined that the similarity of the searched object is not greater than a set value through the feature comparison process, it is determined that the searched component is not the first component 160 to be replaced. Next, the second object stored in the second database is used to replace the first object 170 in the dynamic image being played.

以下所述即为利用本发明的方法取代动态影像内的部分元件的其中之一实施例,其中此动态影像为一动态影像档或是即时动态影像,但并不限制本发明的保护范围。The following description is one of the embodiments of using the method of the present invention to replace some components in the dynamic image, wherein the dynamic image is a dynamic image file or real-time dynamic image, but does not limit the protection scope of the present invention.

通常在动态影像内均包含人物的画面,若影片上的人物想要隐藏自己的真实面目而不被观赏者辨识出影片上的人物的真实面目,则通常采用传统的方法,对影片上的人物的脸孔施以雾状的处理,以使使用者无法辨识出影片上的人物的真实面貌。但是此种雾状处理的方式较容易被人所破解,也较易破坏影像整体的美观,因此必须采用本发明的方法以确实保护影片中的人物的隐私,并增加娱乐及美观的效果。Generally, pictures of characters are included in the dynamic image. If the characters on the film want to hide their true faces without being recognized by the viewers, traditional methods are usually used to treat the characters on the film. Faces are fogged so that users cannot recognize the real faces of the people on the video. But this kind of mist treatment is easier to be deciphered by people, and it is also easier to destroy the overall beauty of the image. Therefore, the method of the present invention must be used to protect the privacy of the characters in the film, and increase the entertainment and aesthetic effects.

首先必须先在电脑上安装本发明的方法所提供的程序并驱动此程序。接下来使用者可利用使用者介面选择由储存在第二资料库内的多种图形中选出一种图案或是多种图案,以取代动态影像中的人物的脸孔。储存在第二资料库内的多种图形,通常为动物的脸孔,以增加播出动态影像时的娱乐效果。若动态影像档案中只出现单一之人物,则使用者只需在使用者介面上选择一图像即可,以在后续的步骤中取代动态影像中的单一人物的脸孔。若动态影像中出现多个不同的人物,则使用者可在使用者介面上依序选择不同的图像,以在后续的步骤中取代动态影像中的不同人物的脸孔。使用者也可只在使用者介面上选择一种图像,但是在播出动态影像时,不同的人物的脸孔将会以同一个图像取代。Firstly, the program provided by the method of the present invention must be installed and driven on the computer. Next, the user can use the user interface to select one or more patterns from the various patterns stored in the second database to replace the faces of the characters in the dynamic image. A variety of graphics stored in the second database, usually the faces of animals, to increase the entertainment effect when broadcasting dynamic images. If only a single character appears in the dynamic image file, the user only needs to select an image on the user interface to replace the face of the single character in the dynamic image in subsequent steps. If multiple different characters appear in the dynamic image, the user can sequentially select different images on the user interface to replace faces of different characters in the dynamic image in subsequent steps. The user can also select only one image on the user interface, but when the moving image is played, the faces of different characters will be replaced by the same image.

接下来使用者可开始播放动态影像,其中此动态影像为一动态影像档或是即时动态影像。在播放影像的同时,本发明的程序即由第一资料库内所储存的特征来搜寻人的脸孔并进行一相似度的判断。因为本实施例为取代动态影像内的人物的脸孔,因此在第一资料库内所储存的特征值均为判断人类脸孔的特征值,如:肤色、五官的位置及眼睛的颜色等等。在搜寻脸孔的过程中,必须同时进行一相似度的判断。当动态影像内的物件和第一资料库内所储存的特征值比较过后所得到的一相似度大于一设定值时,则判断此物件为人的脸孔。若动态影像内的物件和第一资料库内所储存的特征值比较过后所得到的一相似度不大于一设定值时,则判断此物件不为人的脸孔。若动态影像内有多个不同的脸孔则由各不同脸孔在动态影像内所移动的不同的轨迹而分辨其为不同的脸孔,并依序编号以使各不同的脸孔对应到使用者在使用者介面上所选择的不同的图像。Next, the user can start to play the dynamic image, wherein the dynamic image is a dynamic image file or a real-time dynamic image. While playing the video, the program of the present invention uses the features stored in the first database to search for human faces and perform a similarity judgment. Because this embodiment replaces the faces of the characters in the dynamic image, the feature values stored in the first database are all feature values for judging human faces, such as skin color, the position of facial features, and the color of eyes, etc. . In the process of searching faces, a similarity judgment must be performed at the same time. When the similarity obtained after comparing the object in the dynamic image with the feature value stored in the first database is greater than a set value, it is determined that the object is a human face. If the similarity obtained after comparing the object in the dynamic image with the feature value stored in the first database is not greater than a set value, it is determined that the object is not a human face. If there are many different faces in the dynamic image, they are distinguished as different faces by the different trajectories of the different faces moving in the dynamic image, and they are numbered in sequence so that each different face corresponds to the user. or a different image selected on the user interface.

当辨认出动态影像内的人物脸孔之后,随即将使用者于使用者介面上所选择的图像取代人物的脸孔。观赏者在屏幕上只可看见一替代人物脸孔的图像,并无法得知动态影像内的人物的真实面貌。随着使用者不同的需求,使用者可利用本发明的程序的使用者介面在播放动态影像的时候撷取其中的一张图像为静态影像。在动态影像档播放完毕之后,使用者更可利用使用者介面选择是否以图像取代人物脸孔的动态影像档覆盖原有的动态影像档,或是将以图像取代人物脸孔的动态影像档另外储存为一个新的动态影像档。在利用第二资料库的图像取代动态影像内的人物脸孔时,取代脸孔的图像可随着脸孔在动态影像内的尺寸而放大或是缩小,因此并不容易让观赏者得知动态影像内的人物的真实面目。当以图像取代人物脸孔的档案储存为一新的档案后,动态影像档案内的人物的脸孔已被图像所取代,因此其他人无法由其他的方式而得知动态影像档案内的人物的真实面貌,因此可以保护影片中人物的隐私,并增加影片的娱乐效果。After recognizing the face of the person in the dynamic image, the image selected by the user on the user interface is replaced with the face of the person. Viewers can only see an image that replaces the face of the person on the screen, and cannot know the real appearance of the person in the dynamic image. According to the different needs of the users, the user can use the user interface of the program of the present invention to capture one of the images as a static image when playing the dynamic images. After the dynamic image file is played, the user can also use the user interface to choose whether to replace the original dynamic image file with the dynamic image file with the image replacing the face of the person, or to replace the dynamic image file with the image instead of the face of the person separately. Save as a new motion picture file. When using the image from the second database to replace the face of the person in the video, the image replacing the face can be enlarged or reduced according to the size of the face in the video, so it is not easy for the viewer to know the dynamics The real face of the person in the image. When a file that replaces a person's face with an image is saved as a new file, the face of the person in the moving image file has been replaced by an image, so other people cannot know the identity of the person in the moving image file in other ways Therefore, the privacy of the characters in the film can be protected and the entertainment effect of the film can be increased.

参照图2所示,此为本发明的第二流程图。此第二流程图为利用本发明的方法以图像取代动态影像内的人物脸孔的流程图。首先必须先在电脑上安装本发明的方法所提供的程序并驱动此程序210。接下来使用者可利用使用者介面选择由储存在第二资料库内的多种图形中选出一种图案或是多种图案220,以取代动态影像中的人物的脸孔。接下来使用者可开始播放动态影像,其中此动态影像为一动态影像档或是即时动态影像230。在播放影像的同时,本发明的程序即由第一资料库内所储存的特征来搜寻人的脸孔并进行一相似度的判断240。当动态影像内的物件和第一资料库内所储存的特征值比较过后所得到的一相似度大于一设定值时,则判断此物件为人的脸孔250。若动态影像内的物件和第一资料库内所储存的特征值比较过后所得到的一相似度不大于一设定值时,则判断此物件不为人的脸孔260。若动态影像内有多个不同的脸孔则由各不同脸孔在动态影像内所移动的不同的轨迹而分辨其为不同的脸孔,并依序编号以使各不同的脸孔对应到使用者在使用者介面上所选择的不同的图像。当辨认出动态影像内的人物脸孔之后,随即将使用者于使用者介面上所选择的图像取代人物的脸孔270。Referring to FIG. 2 , this is the second flowchart of the present invention. The second flow chart is a flow chart of using the method of the present invention to replace the faces of people in the dynamic image with images. Firstly, the program provided by the method of the present invention must be installed on the computer and the program 210 must be driven. Next, the user can use the user interface to select a pattern or patterns 220 from various patterns stored in the second database to replace the faces of the characters in the dynamic image. Next, the user can start to play the dynamic image, wherein the dynamic image is a dynamic image file or a real-time dynamic image 230 . While playing the video, the program of the present invention searches for human faces from the features stored in the first database and performs a similarity judgment 240 . When the similarity between the object in the dynamic image and the feature value stored in the first database is greater than a set value, it is determined that the object is a human face 250 . If the similarity between the object in the dynamic image and the feature value stored in the first database is not greater than a set value, it is determined that the object is not a human face 260 . If there are many different faces in the dynamic image, they are distinguished as different faces by the different trajectories of the different faces moving in the dynamic image, and they are numbered in sequence so that each different face corresponds to the user. or a different image selected on the user interface. After the face of the person in the dynamic image is recognized, the image selected by the user on the user interface is then replaced with the face of the person 270 .

综合上述,本发明提供了一种方法,利用资料库中特征的辨识与物件的取代替换即时动态影像档案与动态影像档案内的部分物件,以确实隐蔽即时动态影像档案或是动态影像档案内所欲隐蔽的部分物件。本发明的替换方法为利用一程序以取代即时动态影像与动态影像内的物件,其中上述的程序包含一第一资料库、一第二资料库与一使用者介面(user interface)。第一资料库储存的内容为在动态影像内所欲取代的第一物件的多个特征。第二资料库储存的内容为欲取代第一物件的第二物件。在动态影像播放时,本发明的程序即由第一资料库内的物件特征开始搜寻动态影像内的第一物件并进行一物件特征对比的程序。在物件特征对比的程序中,本发明的程序由一相似度以搜寻动态影像内所包含的第一物件。当所搜寻的物件经由特征对比的程序中判断其相似度大于一设定值时,即判断所搜寻的物件为第一物件。接下来利用第二资料库内所储存的第二物件以取代正在播放中的动态影像内的第一物件。在动态影像播放的过程中,使用者可利用使用者介面撷取一以第二物件取代第一物件的静态影像。当动态影像播放完毕后,使用者可在使用者介面上选择是否要以第二物件的动态影像取代第一物件的动态影像。当欲取代的第一物件为多个第一物件且不同的第一物件需以不同的第二物件取代时,则可利用不同的多个第一物件在动态影像或是即时动态影像内所移动的轨迹来分类,并将同一类的第一物件以相同的第二物件所取代。本发明的方法也可以增加即时动态影像档案与动态影像档案的娱乐效果并充分保护个人的隐私。本发明的方法更可加速影像处理的效果,不仅具有实用功效,并且为前所未见的设计,具有功效性与进步性的增进。To sum up the above, the present invention provides a method to replace real-time dynamic image files and some objects in the dynamic image files by using the identification of features in the database and the replacement of objects, so as to truly conceal the real-time dynamic image files or all objects in the dynamic image files. Part of the object to be hidden. The replacement method of the present invention is to use a program to replace real-time dynamic images and objects in the dynamic images, wherein the above-mentioned program includes a first database, a second database and a user interface (user interface). The content stored in the first database is a plurality of characteristics of the first object to be replaced in the dynamic image. The content stored in the second database is the second object to replace the first object. When the dynamic image is played, the program of the present invention searches for the first object in the dynamic image from the object characteristics in the first database and performs a comparison of object characteristics. In the process of object feature comparison, the program of the present invention searches for the first object contained in the dynamic image by a similarity. When the similarity of the searched object is determined to be greater than a set value through the feature comparison program, it is determined that the searched object is the first object. Next, the second object stored in the second database is used to replace the first object in the dynamic image being played. In the process of playing the dynamic image, the user can use the user interface to capture a static image in which the first object is replaced by the second object. After the dynamic image is played, the user can choose whether to replace the dynamic image of the first object with the dynamic image of the second object on the user interface. When the first object to be replaced is a plurality of first objects and different first objects need to be replaced by different second objects, different first objects can be used to move within the dynamic image or real-time dynamic image trajectories, and replace the first object of the same class with the same second object. The method of the present invention can also increase the entertainment effect of real-time dynamic image files and dynamic image files and fully protect personal privacy. The method of the present invention can accelerate the effect of image processing, not only has practical effect, but also has an unprecedented design, and has improved effect and progress.

以上所述仅为本发明的较佳实施例,此实施例仅用来说明而非用以限定本发明的保护范围。在不脱离本发明的实质内容的范畴内仍可予以变化而加以实施,此等变化应仍属本发明的保护范围。因此,本发明的范畴应由权利要求书所界定。The above description is only a preferred embodiment of the present invention, and this embodiment is only used for illustration rather than limiting the protection scope of the present invention. Changes can still be made and implemented within the scope of not departing from the essence of the present invention, and these changes should still belong to the protection scope of the present invention. Therefore, the scope of the present invention should be defined by the claims.

Claims (16)

1.一种动态影像内的物件的替换方法,其特征在于,包含:1. A replacement method for an object in a dynamic image, characterized in that it comprises: 安装一包含一使用者介面、一第一资料库及一第二资料库的程序于一电脑内并驱动该程序,其中上述的第一资料库包含多个第一物件的多个特征而该第二资料库包含多个第二物件;Install a program including a user interface, a first database and a second database in a computer and drive the program, wherein the above-mentioned first database contains a plurality of characteristics of a plurality of first objects and the second database contains a plurality of characteristics of a plurality of first objects The second database contains a plurality of second objects; 在该使用者介面上选择部分的该多个第一物件与部分的该多个第二物件;selecting a portion of the plurality of first objects and a portion of the plurality of second objects on the user interface; 显示一动态影像于一屏幕上;displaying a moving image on a screen; 利用部分该多个特征与一设定值搜寻并判别该动态影像内的该多个第一物件;searching for and identifying the plurality of first objects in the dynamic image by using part of the plurality of features and a set value; 当该部分多个特征与该动态影像内的多个第三物件比较后所得到的一相似度大于该设定值时,则判定该多个第三物件为部分该多个第一物件;When the similarity obtained after comparing the plurality of features of the part with the plurality of third objects in the dynamic image is greater than the set value, it is determined that the plurality of third objects are part of the plurality of first objects; 当部分该多个特征与该动态影像内的多个第三物件比较后所得到的一相似度不大于该设定值时,则判定该多个第三物件不为部分该多个第一物件;When a similarity obtained after comparing some of the plurality of features with the plurality of third objects in the dynamic image is not greater than the set value, then it is determined that the plurality of third objects are not part of the plurality of first objects ; 以部分该多个第二物件取代部分该多个第一物件,并显示于屏幕上;及replacing part of the plurality of first objects with part of the plurality of second objects and displaying on the screen; and 利用该使用者介面撷取该动态影像的其中之一而为一静态影像。One of the dynamic images is captured by the user interface to be a static image. 2.如权利要求1所述的动态影像内的物件的替换方法,其特征在于,上述的多个特征可为一人类脸孔的特征。2. The method for replacing objects in a dynamic image as claimed in claim 1, wherein the above-mentioned features can be features of a human face. 3.如权利要求2所述的动态影像内的物件的替换方法,其特征在于,上述的人类脸孔的特征可为一肤色。3. The method for replacing objects in a dynamic image as claimed in claim 2, wherein the feature of the above-mentioned human face can be a skin color. 4.如权利要求2所述的动态影像内的物件的替换方法,其特征在于,上述的人类脸孔的特征可为一五官的位置。4. The method for replacing objects in a dynamic image as claimed in claim 2, wherein the above-mentioned feature of a human face can be a position of a five sense organ. 5.如权利要求1所述的动态影像内的物件的替换方法,其特征在于,上述多个第二物件可为动物脸孔的图像。5. The method for replacing objects in a dynamic image as claimed in claim 1, wherein the plurality of second objects are images of animal faces. 6.一种动态影像内的物件的替换方法,其特征在于,包含:6. A method for replacing objects in a dynamic image, comprising: 安装一包含一使用者介面、一第一资料库及一第二资料库的程序于一电脑内并驱动该程序,其中上述的第一资料库包含多个第一物件的多个特征而该第二资料库包含多个第二物件;Install a program including a user interface, a first database and a second database in a computer and drive the program, wherein the above-mentioned first database contains a plurality of characteristics of a plurality of first objects and the second database contains a plurality of characteristics of a plurality of first objects The second database contains a plurality of second objects; 在该使用者介面上选择部分的该多个第二物件;select a portion of the plurality of second objects on the user interface; 显示一第一动态影像于一屏幕上;displaying a first dynamic image on a screen; 利用该多个特征与一设定值搜寻并判别该第一动态影像内的该多个第一物件;searching for and identifying the plurality of first objects in the first dynamic image by using the plurality of features and a set value; 当该多个特征与该第一动态影像内的多个第三物件比较后所得到的一相似度大于一设定值时,则判定该多个第三物件为该多个第一物件;When a similarity obtained after comparing the plurality of features with the plurality of third objects in the first dynamic image is greater than a set value, then determining that the plurality of third objects are the plurality of first objects; 当该多个特征为与该第一动态影像内的多个第三物件比较后所得到的一相似度不大于一设定值时,则判定该多个第三物件不为该多个第一物件;When the multiple features are compared with the multiple third objects in the first dynamic image, the similarity obtained is not greater than a set value, then it is determined that the multiple third objects are not the multiple first dynamic images. object; 利用该多个第一物件于该第一动态影像内的多个不同的移动轨迹判断出不同的该多个第一物件;using multiple different moving trajectories of the multiple first objects in the first dynamic image to determine different multiple first objects; 在该第一动态影像上以该部分多个第二物件取代该多个第一物件而变成一第二动态影像,并将该第二动态影像显示于屏幕上,其中上述的该第二动态影像中,同一该第一物件可以同一该第二物件取代;及Replace the plurality of first objects with the part of the plurality of second objects on the first dynamic image to become a second dynamic image, and display the second dynamic image on the screen, wherein the above-mentioned second dynamic in the image, the same first object may be replaced by the same second object; and 储存该第二动态影像于该电脑中。The second dynamic image is stored in the computer. 7.如权利要求6所述的动态影像内的物件的替换方法,其特征在于,上述的多个特征可为一人类脸孔的特征。7. The method for replacing objects in a dynamic image as claimed in claim 6, wherein the above-mentioned features can be features of a human face. 8.如权利要求7所述的动态影像内的物件的替换方法,其特征在于,上述的人类脸孔的特征可为一肤色。8. The method for replacing objects in a dynamic image as claimed in claim 7, wherein the feature of the above-mentioned human face can be a skin color. 9.如权利要求7所述的动态影像内的物件的替换方法,其特征在于,上述的人类脸孔的特征可为一五官的位置。9. The method for replacing objects in a dynamic image according to claim 7, wherein the above-mentioned feature of a human face can be a position of a five sense organ. 10.如权利要求6所述的动态影像内的物件的替换方法,其特征在于,上述的第二动态影像可取代该第一动态影像。10. The method for replacing objects in a dynamic image as claimed in claim 6, wherein the second dynamic image can replace the first dynamic image. 11.如权利要求6所述的动态影像内的物件的替换方法,其特征在于,上述的第一动态影像可为一即时动态影像。11. The method for replacing objects in a dynamic image as claimed in claim 6, wherein the first dynamic image is a real-time dynamic image. 12.一种动态影像内的物件的替换方法,其特征在于,包含:12. A method for replacing objects in a dynamic image, comprising: 安装一包含一使用者介面、一第一资料库及一第二资料库的程序于一电脑内并驱动该程序,其中上述的第一资料库包含多个人类脸孔的多个特征而该第二资料库包含多个图像;Install a program including a user interface, a first database and a second database in a computer and drive the program, wherein the above-mentioned first database contains a plurality of features of a plurality of human faces and the second database includes a plurality of features of a plurality of human faces. Two databases containing multiple images; 在该使用者介面上选择部分的该多个图像;select a portion of the plurality of images on the user interface; 显示一第一动态影像于一屏幕上;displaying a first dynamic image on a screen; 利用该多个特征与一设定值搜寻并判别一第一动态影像内的该多个人类脸孔;Searching and discriminating the plurality of human faces in a first dynamic image by using the plurality of features and a set value; 当该多个特征与该第一动态影像内的多个物件比较后所得到的一相似度大于一设定值时,则判定该多个物件为该多个人类脸孔;When the similarity obtained after comparing the plurality of features with the plurality of objects in the first dynamic image is greater than a set value, it is determined that the plurality of objects are the plurality of human faces; 当该多个特征与该第一动态影像内的多个物件比较后所得到的一相似度不大于一设定值时,则判定该多个第三物件不为该多个人类脸孔;When the similarity obtained after comparing the multiple features with the multiple objects in the first dynamic image is not greater than a set value, then it is determined that the multiple third objects are not the multiple human faces; 利用该多个人类脸孔于该第一动态影像内的多个不同的移动轨迹判断出不同的该多个人类脸孔;using multiple different moving trajectories of the multiple human faces in the first dynamic image to determine the different multiple human faces; 在该第一动态影像上以该多个图像取代该多个人类脸孔而变成一第二动态影像,并将该第二动态影像显示于屏幕上,其中上述的该第二动态影像中,同一该人类脸孔可以同一该图像取代;及replacing the plurality of human faces with the plurality of images on the first dynamic image to form a second dynamic image, and displaying the second dynamic image on the screen, wherein in the above-mentioned second dynamic image, the same human face may be replaced by the same image; and 储存该第二动态影像于该电脑中。The second dynamic image is stored in the computer. 13.如权利要求12所述的动态影像内的物件的替换方法,其特征在于,上述的人类脸孔的特征可为一肤色。13. The method for replacing objects in a dynamic image as claimed in claim 12, wherein the feature of the above-mentioned human face can be a skin color. 14.如权利要求12所述的动态影像内的物件的替换方法,其特征在于,上述的人类脸孔的特征可为一五官的位置。14. The method for replacing objects in a dynamic image as claimed in claim 12, wherein the above-mentioned feature of a human face can be a position of a five sense organ. 15.如权利要求12所述的动态影像内的物件的替换方法,其特征在于,上述的第二动态影像可取代该第一动态影像。15. The method for replacing objects in a dynamic image as claimed in claim 12, wherein the second dynamic image can replace the first dynamic image. 16.如权利要求12所述的动态影像内的物件的替换方法,其特征在于,上述的第一动态影像可为一即时动态影像。16. The method for replacing objects in a dynamic image as claimed in claim 12, wherein the first dynamic image is a real-time dynamic image.
CN 02119710 2002-05-10 2002-05-10 How to replace objects in dynamic images Pending CN1456970A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 02119710 CN1456970A (en) 2002-05-10 2002-05-10 How to replace objects in dynamic images

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 02119710 CN1456970A (en) 2002-05-10 2002-05-10 How to replace objects in dynamic images

Publications (1)

Publication Number Publication Date
CN1456970A true CN1456970A (en) 2003-11-19

Family

ID=29410424

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 02119710 Pending CN1456970A (en) 2002-05-10 2002-05-10 How to replace objects in dynamic images

Country Status (1)

Country Link
CN (1) CN1456970A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106657812A (en) * 2008-03-03 2017-05-10 威智伦分析公司 Dynamic object classification

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106657812A (en) * 2008-03-03 2017-05-10 威智伦分析公司 Dynamic object classification
US10417493B2 (en) 2008-03-03 2019-09-17 Avigilon Analytics Corporation Video object classification with object size calibration
CN106657812B (en) * 2008-03-03 2020-04-03 威智伦分析公司 A camera system and dynamic object classification method
US10699115B2 (en) 2008-03-03 2020-06-30 Avigilon Analytics Corporation Video object classification with object size calibration

Similar Documents

Publication Publication Date Title
KR101348521B1 (en) Personalizing a video
US6393147B2 (en) Color region based recognition of unidentified objects
US9762775B2 (en) Method for producing a blended video sequence
TW544634B (en) Thumbnail sequence generation system and method
CN101615389B (en) Electronic apparatus, and video content editing method
US7034833B2 (en) Animated photographs
CN108307229B (en) Video and audio data processing method and device
US9576610B2 (en) Systems and methods of detecting significant faces in video streams
US7904815B2 (en) Content-based dynamic photo-to-video methods and apparatuses
US20030051255A1 (en) Object customization and presentation system
CN1522425A (en) Method and device for superimposing user image on original image
CN107111866A (en) Method and apparatus for generating extrapolated image based on object detection
JP2003204541A (en) Video processing method and video processing apparatus
Rea et al. Multimodal periodicity analysis for illicit content detection in videos
US20080068397A1 (en) Emotion-Based Digital Video Alteration
JP5775814B2 (en) Apparatus and method for adjusting audiovisual system to viewer's attention level
US8184869B2 (en) Digital image enhancement
US20030206654A1 (en) Replacing method of an object in a dynamic image
US20110304644A1 (en) Electronic apparatus and image display method
JPH099202A (en) Index generation method, index generation device, indexing device, indexing method, video minutes generation method, frame editing method and frame editing device
Teodosio et al. Salient stills
CN1456970A (en) How to replace objects in dynamic images
Peker et al. An extended framework for adaptive playback-based video summarization
WO2009044351A1 (en) Generation of image data summarizing a sequence of video frames
TWI225225B (en) Replacing method of an object in a dynamic image

Legal Events

Date Code Title Description
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication