TWI254569B - System and method of network interactive-type image processing - Google Patents

Abstract

The present invention relates to a system and method of network interactive-type image processing, and especially relates to a system and method to adjust the portrait image by utilizing the network features. Through the image-processing module stored at different ends of network and the adjusting template database of portrait image, the portrait image data is adjusted. The adjusting template database of portrait image comprises plural portrait-adjusting data. The plural portrait-adjusting data can be generated by computer software or by manual adjustment, and comprise the adjusting data to remove the defect of portrait skin, the data to adjust the distribution ratio of five facial organs, and the data to adjust the facial expression such that they can be used to modify the facial expression of the portrait for generating the effect of different portrait image.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system and method for network interactive image processing, particularly relating to a system and method for utilizing network characteristics to adjust portrait images, which are stored through A system and method for adjusting a portrait image at a plurality of different image processing modules at different ends of the network and portrait adjustment parameters in a portrait image adjustment template database. [Prior Art] With the widespread use of the Internet, there are more and more ways to process data through the Internet, including static image data and dynamic Processing of video data and sounds and texts. At the same time, with the popularity of wireless terminal devices (such as mobile phones and personal digital assistants), the need to process information content using wireless terminal devices is increasing. Although users of wireless terminal devices can process the content of information through wireless terminal devices, some problems still have to be overcome: Brother I, " w π 热外~1 with six-one computer in data capacity, data processing There is a straight line in speed, display standard and memory capacity. I: : Features and limitations ' 0 This data processing method for general computer use can hardly be executed and used on wireless terminal devices. Second, the speed of data processing by wireless terminal devices is slower than that of a general desktop j-type computer, and it takes a long time to process data. In particular, if you include a large number of pictures and texts in the processing, the user should use

Page 5 1254569

The wireless terminal information processing is limited by the development of processing tools. The end device of the product is this, the processing tool. In the same way, wirelessly, the addition and popularization of the genre, the mainstream of the three: the news will have a huge problem. The second ' < user will use the wireless terminal device as a secondary Beixun Valley can be processed through the wireless terminal device is the tenth, for the processing of specific information content, such as image data,,; Unsatisfactory. Digital images are easier to use as image processing software than traditional photos, and they are more flexible in application. With the digital imaging technology, the consumer market's preference for digital imaging products and related industries have also flourished, becoming digital digital imaging technology from the laboratory into the professional field, and then into the public life from the professional field, for the application In addition to the functions and quality, the requirements of tools are easy to learn and use, so the wisdom and automation of the tools become an important source of competitiveness for the products. It is also the focus and trend of the new generation of digital shadow tools. ~ The traditional methods used to deal with the facial features and expressions of portrait digital images are as follows: ^ Use image processing software (for example: P h 〇t 〇sh ο p ) Manual finishing · Various brushes for image processing, Filters and deformation effects, one by one ♦ decorated portrait photos 'The main disadvantage is that you need to repeatedly operate complex image processing tools, users must be full of skilled image processing tools = and = professional literacy professionals. All the steps are manual and ^re-reserved, so each photo needs to be processed individually. μ / use the original photo to interpolate with another photo (

1254569 V. Invention description (3) M〇rph) to generate a new intermediate picture (for example, Morpheus s^ftware), the main disadvantage is that there must be two photos as 'if the two photos are not the same person, Then the process of deformation is less and less like myself. As described above, how to process digital images on the network is a complicated problem, and many applications are in use in response to the above-mentioned situation. The present invention provides a system and method for network interactive processing. Not only can it improve the shortcomings of the traditional complex shadow tools, but also can reuse the set values, and can be used on the wireless terminal equipment and the Internet, completely overcome the lack of problems and problems, and need to be processed on the Internet. Image users are undoubtedly a convenient solution. The original element makes the image of the image face the image like the processing source is restricted. The invention is based on the image of the image of the invention. Like the processing skills, the main method and method system and the number of image points, the database is more than the portrait, using the special method of the purpose, when the module image contains the complex and the parameters, it is to provide a kind of The net is a kind of arrangement and combination of the characteristics of the symbolic pen portrait change through the storage and portrait image adjustment, and the simple adjustment method uses the net to store the network adjustment system and the square parameter material, so as to constitute the routine interactive road. This feature data is different from this data method. This information is used to improve the person's different use of the replacement image to adjust the person's portrait image in the plural library of the human end. For example: the shadow image professional image professional image

1254569 V. INSTRUCTIONS (4) The secondary object of the present invention is to include a system and method for illuminating the image processing module and the portrait image of the moving image at the same time == multiple plural parameters , the system of portrait image adjustment; the image can be used directly through the selection, when accelerating = professional knowledge like video photos or computer graphics can be used. 'There is no need to have art. Another object of the present invention is to provide a system and a <RTI ID=0.0>>>""""" In the portrait image adjustment database, the human model database contains a plurality of portraits ^' too [, the portrait adjustment adjustment model is generated by computer software or == pen portrait image has the adjustment parameters of the portrait skin fatigue, five; Zhou is included to produce different m:: effects; the portrait expression is appropriately modified, and the purpose is to provide a system and method for network interaction ηβ theory, especially with J-style "! The system and method of image, wherein the characteristics of the invention are combined with the image of the human face by the technique of adjusting the lens, the local deformation technique and the portrait muscle method, wherein the filter technique is used for C, etc. to adjust the skin: Smoother and healthier, such as elimination: =: like = so that the two deformation techniques are used to adjust the shape of the face, the proportion of the five senses and the shape of the court =: Meat: Page 8 1254569 V. Description of the invention (5) The technique is based on the human expression, To present Now adjust the parameter value to use the storage sleeve for the portrait image. The system and method of the present invention is like the image system and a different image image adjustment parameter, and the data processing of different devices stored by using different image processing systems is processed. The efficiency and method of the image. The anatomical relationship of the personality changes, in the adjustment of a 'substantial improvement of the purpose, in particular, there are methods, the system module and the portrait image module and the portrait of the network does not count resources. In order to provide the use of the structure, through the muscles through the above-mentioned image data processing in the image database to provide a kind of gateway through the image of the image of the whole image of the same end, the different people to facilitate the use of the network to adjust the network and The interactive image processing of the data of the full-fledged model is so that the effect of the time and the interactive road characteristics can be adjusted according to the different forms of the database. The database is used to adjust the data on the network. Obtain the speed after selection. The image is adjusted in the plural of the person. The system of the road [Embodiment] In order to make the Responsible Review Committee have a better understanding and understanding of the architectural features and the achieved effects of the present invention, please provide a more detailed description of the following: Please refer to FIG. 1 , which is a system architecture diagram of the present invention. As shown in FIG. 1A, the present invention provides a network interactive video media processing system including a media providing unit (media pr〇vider) 〇 1 A source for providing media, an object feature identification unit (Feature p〇int identif ier) 02 for identifying and locating the object feature, a network 〇 3, page 9 V, invention description (6) a media generation unit (media generator) 〇4 is used to adjust the template information according to the object and the selected object, and generate the adjusted output material, a template database 〇5 for storing a plurality of preset object adjustments. The template, the different object adjustment 2 also includes different preset object adjustment parameters. The media providing unit 本1 of the present invention first provides a media image, and the source of the smock image can be located in a network 〇3 (which can be an internet network), and the media material is in the present invention. After the feature identification of the object 兀02, the relevant features are positioned. The identification data :: :: ΐ ” will be transmitted to the network through the network 03 另 3 another - end 、 , - generate early 兀 04, the media production unit 〇 4 will adjust the template information according to the selected object , Feature Bay 05 selects the required object adjustment template to generate the second adjustment. Please refer to Figure 1 for the data base, as shown in the figure: a data 〇 8. Inter-type image media processing system It can also include a media-species, material and a codec unit ((:01){:(:)〇7 using the moxibustion of the media file to adjust the template information in the network of the anti-Temple information and objects connected to the J-end solution 'Or can be the object feature information J shell: the two developed end coded and embedded in the original media file:::: 扼 lose, and decode and restore at the receiving end.

A Basically, the present invention is a network mutual system and method, and the operation method thereof can be =~ like the media processing system h. FIG. 2A is the _~ sample of the present invention, first refer to FIG. #网络互动式影像媒体处理系统1254569 V. Description of the Invention (7) A flow chart of one embodiment. As shown in FIG. 2, the main steps of the present invention include at least: and method six, eight v--Guo. The main steps of the embodiment include at least: Step 21A: Loading a portrait media through a media providing unit Step 22A: identifying and locating the portrait feature of the portrait media image via an object feature recognition unit; Step 23A: inputting the recognized and positioned image media image to a media generation unit and one of the other ends of the network Downloading a portrait image adjustment template from the media image adjustment database to the media generation unit% sensitive use* Each person image media image adjustment parameter value included in the portrait image adjustment template is adjusted to the portrait media image; = step 24A: output The adjusted person is like a media image. 2B, FIG. 2B is a flowchart of one embodiment of the image media processing system and method according to the present invention. As shown in FIG. 2B, the main steps of the embodiment of the present invention include: loading a portrait media image file through a media providing unit via a network end or a media storage unit of the network 2; ST: identifying and locating the portrait feature of the portrait media image through an object feature recognition unit; to _=2A3B: inputting the recognized and positioned image media image template list = from the other end of the network - portrait media image adjustment Fen, the community downloads a portrait image adjustment template to the media production unit, and then: estimate the image image of each person image included in the portrait image adjustment template to the portrait media image;

1254569 V. INSTRUCTIONS (8) Step 24B: Please refer to the media processing system shown in Figure 3, Step 31: Case; Step 32: Media Image Person Step 33: A media generated in a single database, Apply the person to adjust the parameter value to step 34: One of the network ends is modulated to generate a single image media step; the other end of the step is like the media integer, the image is adjusted to 3 5 · or 36: 〇 Refer to the processing system The output of the adjustment is shown in Figure 3, and Figure 3 is the portrait media image of this issue.统和方法#中的_] 二-network interactive video through a media delivery system contains at least: 疋 入 一 一 一 媒体 媒体 媒体 媒体

ί: The object identification unit identifies and locates the portrait input and recognizes the location and the I and the other end of the network: the portrait media image of the person to the image-image image adjustment; the image media image adjustment image image adjustment In the template, the media generating unit is the person media image; <3 each person media image transmission and identification and positioning of the media production unit and the media image, the portrait media application #a push a portrait image adjustment template to the media ΛΛ Adjust the positive parameter values of each person included in the template to the portrait media image. • Turn the adjusted image of the person like the media image on the network. The adjusted image of the person is in the network image 4, Figure 4 A flowchart of still another embodiment of a network interaction, prior and method of the present invention. As shown

Page 12 1254569

1254569 V. Description of Invention (10) Code and download a portrait image adjustment template from the portrait media image adjustment model at the second end of the network to the media generation unit order and included in the Putian portrait image adjustment template Each person like media shadow = to the portrait media image; η positive parameter value step 55 · output the adjusted media image at the third end of the network. The present invention can be understood from the above various embodiments, and a method of image media processing of the present invention includes a plurality of different implementations. Each work unit in the system can be restricted by different network resources. At different ends of the network, for example, the device at one end of the network may be a PDA, and its data storage and processing capabilities are limited (for example: ', ' Insufficient memory, insufficient storage space, insufficient CPU performance, etc., so the direct 1 port '^ must use the object recognition unit of ft easy to identify and locate the portrait feature of the person, and need to store a large amount of data or process data. Worksheets -: Portrait media image adjustment template database and media:: Set: the other end of the network, download and use when needed. The media data output at the same time can also be output at different ends of the network as needed. The object feature identification unit in the object feature recognition unit system in the system of the present invention firstly identifies and locates the input portrait media shirt image to determine the person's portrait feature, and analyzes the image based on the portrait image feature. Calculate the position of the official character of the five senses, and then construct a structure diagram of the second sign according to the set of features of the five features. The portrait image is adjusted according to one of the portrait images of the present invention according to the five = &

1254569

Take a picture of the adjustment, the media image to modify the body image to be able to re-do friends. When the portrait image of the image is adjusted, the skin color and the image are applied in tandem with the portrait image. The image is adjusted in this document, and the person like the five light, the mouth and the invention system are subjected to various embodiments. Like the above-mentioned model modification, the invention uses the adjustment template to use the adjustment template, which requires the human portrait image of the invention to speak between the sample books at the same time, and the library can be more interesting to the specific position angle tick.

The adjustment parameters of the adjustment template are applied to the portrait image. After that, the modified person is processed in the next step, for example, using a mobile phone to transmit to a remotely set pre-set portrait image adjustment model, which can be adjusted for different treatments, for example, in addition to being able to pass through a preset portrait. In addition to the shape of the image adjustment template, the size and skin color of the skin can be transmitted through the pre-set portrait image. In the image processing mirror technology, the skin defect and the adjustment system in the image are partially eliminated, and after the facial features are identified, the sample parameter values are adjusted through the portrait image in the human database of the present invention, and the processing is the same as the partial filter, which can eliminate Points and lines while maintaining the clarity of the facial features. For example, the image adjustment template database may include a template of a combination of changes in state, for example, "the facial features of the facial features are presented in an animated manner, which uses continuous interpolation of portrait images, plus a voice-to-mouth, to make a portrait There is a vivid expression. The portrait image adjustment model of the present invention includes a "dynamic tandem symbol effect" which is presented in an animated manner, and the dynamic comic symbol is drawn on the display, for example, when the speech is accompanied by a tearful water like a blink of an eye. , the blue veins of the forehead, etc., can be used to exaggerate the feelings. As described above, the portrait image adjustment model of the present invention

Page 15 1254569

1254569 Brief description of the drawing [Simplified description of the drawing] Figure 1 is a system architecture diagram of the present invention. Figure 1B is a system architecture diagram of another embodiment of the present invention. Figure 2A is a flow chart of a first embodiment of the present invention. Figure 2B is another flow chart of the first embodiment of the present invention. Figure 3 is a flow chart of a second embodiment of the present invention. Figure 4 is a flow chart of a third embodiment of the present invention. Figure 5 is a flow chart of a fourth embodiment of the present invention. Description of the figure: 01- - Media providing unit 02 - Object feature identifying unit 0 3 - - Network 04 - Media generating unit 05 - Adjusting template database 06 - Media storage unit 0 7 - Code decoding unit 08 - Output data 21, 22, 23, 24, 31, 32, 33, 34, 35, 36, 41, 42, 43, 44, 45, 51, 52, 53, 54, 55 - process steps

Page 17

Claims (1)

The invention has a patent range of 1 - 1 · a network failure 7? 4 - M ^ type image media processing system, comprising: a / body providing unit for providing media objects; a feature identification unit for identifying and locating the Media information; 丨丁心何路' is used to transmit the media object; —:Ϊ Ϊ ί 料 料 ' ' to store a plurality of pre-set media: f ^ this, the different media object adjustment template includes different W no a and then media object adjustment parameters; a Ξ ί ί 2 yuan, used to adjust the model based on the characteristics of the media object and the selected media object selected from the data 4, the media after 5 weeks of production object. 2. = One of the processing systems described in the scope of the patent application, including: · J 吟 立 立 立 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象 象And in the network to solve the network transmission, the code and the end of the original media ^ = : = = : the other end of the decoding and reproduction and transmission, and a media storage in the network Unit for yarn six & Second, the pieces. Exist in the original or the adjusted media 3. For example, in the scope of patent application, like the media processing system, j:; the above-mentioned type of network interactive image, where the media object can be - portrait image 1^54569曰修(事) is replacing page—1111 says ^3110053·^ _^j Day correction 6. Patent application scope - " ^ '1 4 · If you apply for patent scope item 1 or 2 The network interactive image media processing system, wherein the media object feature information may be a feature of a relative position of a person or a feature of a portrait skin quality or any combination of the two. 5. A network interactive image media processing system according to claim 1 or 2, wherein the preset media object adjustment number is a parameter of a personlike expression, and a person like a facial features ratio Parameters, parameters like the texture of the skin texture, the number of bright contrasts of a person's skin ^ the parameters of a person's shades of light or any combination of the five. ??? 网路 〆 · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · 预先 预先 预先 预先 预先 预先 预先 预先 预先 预先 预先 预先 预先 预先 预先 预先 预先Or _ = number of 昼 昼 昼 范 效果 效果 效果 或 或 或 或 · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · The processing system, wherein the network can be an Internet-based Internet device, such as the network image processing system of claim 1 or 2, wherein the network can be a Wireless LAN (9) · A network interactive video media processing method, the steps of which include: (a) loading a media object through a media providing unit; (b) An object feature located at either end of the network, such as identifying and locating the characteristics of the media object; " Page 19 1254569; /- ":h ' Case number... VI. Patent application scope (c) • Input the identified and positioned media object to a media generation unit located at the first end of the network and The second end of the network, a media image adjustment template database, downloads an image adjustment r ^ to the media generation list & in the text; (d) · applies the media image adjustments included in the image adjustment template Parameter values to the media object; and ^ (e) outputting the adjusted media object to the first end of the network. i 〇 · A network interactive video medium according to claim 9 of the invention, wherein the media generation unit in step (c) is also located at the second end of the network. 11. The network interactive video media processing method according to claim 9, wherein the adjusted media object can also be rotated at the first end of the network. The network interactive video media processing method of claim 9, wherein the media generating unit in step (c) is located at the second end of the network. The network interactive video media processing method according to claim 12, wherein the adjusted media object can be output to the second end of the network and then output to the first of the network. end. A network interactive video media processing method according to claim 9, wherein in step (c), the media generating unit can be located at a second end of the network. The method for processing a network interactive video medium according to the item of claim 14, wherein the adjusted media object can be output to the network file 鸩 031160 & 1 1 545 ^ 9 曰 repair (〆) replacement page * ♦ f · , , Π 0 1 1 οηπη VI. Patent application scope The third end of the road. A method of network interactive shadow processing according to claim 9, wherein one of the _ terminals of the network is further used, and the decoding unit encodes the media object to be transmitted or A method for processing a network interactive video medium according to claim 9, wherein the media object can be a portrait image. The method of processing a network interactive video medium according to claim 9, wherein the media object feature may be a feature of a relative position of a portrait or a feature of a portrait skin quality or a relationship between the two. Tone combination. A network interactive video media processing method according to claim 9, wherein the image adjustment template may be a combination of a portrait facial expression parameter or a portrait facial feature proportional parameter or both Tone combination. The invention relates to a network interactive video media processing method according to claim 9, wherein the image adjustment template can be a combination of a human skin texture texture parameter, a combination of a human skin bright contrast parameter or a portrait. A combination of color shading parameters or any combination of the three. 2. The network interactive video media processing method according to claim 9, wherein the image adjustment template can be a dynamic series of portrait image change combination templates or a dynamic series comic symbol effect combination. A template or any combination of the ones. 2. The network interactive video media processing method according to claim 9, wherein the media object in step (a) can be located in a network. Page 21 (2545 hard repair (/) is replaced by: 丨 • case sparse (eight) 丨 1CT5:; _Ά 曰 Amendment six, one of the first end of the patent application road or stored in one of the second end of the network Media storage unit. ΙϋΒ page 22