CN106778449B - Object identification method of dynamic image and interactive film establishment method for automatically capturing target image - Google Patents
Object identification method of dynamic image and interactive film establishment method for automatically capturing target image Download PDFInfo
- Publication number
- CN106778449B CN106778449B CN201510818386.1A CN201510818386A CN106778449B CN 106778449 B CN106778449 B CN 106778449B CN 201510818386 A CN201510818386 A CN 201510818386A CN 106778449 B CN106778449 B CN 106778449B
- Authority
- CN
- China
- Prior art keywords
- image
- identification system
- target image
- target
- dynamic
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Abstract
An object identification method of dynamic image uses an identification system as a tool, mainly obtains a dynamic image and an image, then carries on edge detection to the image to decide a target object, then builds a model according to the image characteristic of the target object, captures n key pictures in the image sequence of the dynamic image, finally, compares whether there is an object in the key pictures conforming to the model, if there is an object, captures the key picture as a target image. Therefore, the method can be completely different from the prior art, can immediately identify the required objects aiming at any unexpected dynamic images under the conditions of not needing machine learning time and not needing to build a database, can break through the limitation that the prior art can only identify 2D images, has high identification speed and accuracy, and can be widely applied to the fields of daily life, business and the like.
Description
Technical Field
The present invention relates to an object identification method, and more particularly, to an object identification method for dynamic images and an interactive movie creation method for automatically capturing a target image.
Background
The invention discloses a method and a device for recommending network commodities disclosed in Taiwan patent No. 201508671, which mainly aims at a commodity picture to be inquired including a main commodity by image recognition technology to find a commodity picture to be recommended, wherein the commodity picture has commonality with the commodity picture to be inquired. Therefore, similar commodities can be found out from a large number of commodities in online shopping.
However, the current image recognition technology in the market is as disclosed in the aforementioned 201508671, only static 2D pictures can be recognized, and a database must be established to store recommended commodity pictures and machine learning time (machine learning) before the images can be compared with the commodity pictures to be queried, so that the larger the database is, the more the storage space is consumed, the longer the comparison time is affected, and the disadvantages of management time and manpower waste are caused.
Disclosure of Invention
The invention aims to provide an object identification method of a dynamic image which can identify the dynamic image at a high speed and an interactive film establishment method for automatically capturing a target image.
The invention relates to an object identification method of a dynamic image, which takes an identification system as a tool and comprises the following steps:
step a: the identification system acquires a dynamic image and an image.
Step b: the recognition system performs edge detection on the image to determine a target object.
Step c: the identification system establishes a model according to the image characteristics of the target object.
Step d: the recognition system captures n key frames in the image sequence of the dynamic image, wherein the initial value of n is 1.
Step e: and (e) the identification system compares whether an object which is consistent with the model exists in the key picture, if so, the step f is carried out, and if not, the step g is carried out.
Step f: the identification system captures the key picture as a target image.
Step g: the identification system makes n equal to n +1 and returns to step e.
In the method for identifying an object in a dynamic image according to the present invention, the identification system in step f can simultaneously obtain the time point of the target image appearing in the image sequence of the dynamic image.
In the method for identifying an object of a dynamic image, the model established by the identification system in the step c is a three-dimensional model, and the identification system in the step e can compare whether an object which is consistent with any angle of the model exists in the key picture.
The step e of the object identification method of dynamic image of the invention comprises
Step e-1: the identification system compares whether an object with the similarity of 50-100% with the contour line exists in the key picture according to the contour line of the three-dimensional model, if so, the step e-2 is carried out, if not, the step g is carried out,
step e-2: the recognition system captures the image feature of the object in the key picture,
step e-3: the identification system compares whether the image characteristics of the object in the key picture are consistent with any angle of the three-dimensional model, if so, the step f is carried out, and if not, the step g is carried out.
In the method for identifying an object in a dynamic image according to the present invention, n has a maximum value, and step g includes step g-1: the identification system judges whether n is equal to the maximum value, if yes, the step g-2 is carried out: the identification system stops identification, if not, the step g-3 is carried out: the identification system makes n equal to n +1 and returns to step e.
The invention relates to an object identification method of a dynamic image, wherein the dynamic image and the image come from any storage medium connected with an identification system.
The object identification method of the dynamic image comprises an image acquisition device for acquiring the dynamic image, wherein the dynamic image is a real-time picture acquired by the image acquisition device on site.
The invention relates to an interactive film establishing method for automatically intercepting a target image, which comprises the following steps:
step h after step f: the identification system combines the dynamic image and the selected target image into an interactive film.
Step i: the identification system issues an online video playing interface, and the online video playing interface comprises a first frame for browsing the dynamic images and a second frame for browsing the target images.
The invention relates to an interactive film establishing method for automatically intercepting a target image, which is characterized in that an identification system embeds an online video-audio playing interface into one of a webpage and an application program through a network.
The invention relates to a method for establishing an interactive film capable of automatically intercepting a target image, which is characterized in that the target image is sequentially displayed along with the time axis of a dynamic image in a first frame, and one display mode of the target image and the time axis of the dynamic image in the first frame is synchronously displayed in a second frame.
The interactive film establishing method for automatically intercepting the target image further comprises the following steps j-1 before and after step f: the identification system presents all target images, step j-2: the identification system presents a set layout preloaded with the dynamic image and the target image according to a trigger message of any one target image selected by a point, and the step j-3: the identification system sets the content attribute related to the target image according to the condition input by the setting layout, wherein the content attribute at least comprises a connection website.
The content attribute of the interactive film establishing method for automatically intercepting the target image further comprises name and price.
The invention relates to an interactive film establishing method for automatically intercepting a target image, which comprises a first electronic device and a second electronic device connected with the first electronic device, wherein the second electronic device transmits at least one instruction through an application program, so that the first electronic device executes the steps according to the instruction.
The invention has the beneficial effects that: the method can be completely different from the prior art, can immediately identify the required objects aiming at any unexpected dynamic images under the conditions of not needing machine learning time and not needing to build a database, can break through the limitation that the prior art can only identify 2D images, has high identification speed and accuracy, and can be widely applied to the fields of daily life, business and the like.
Drawings
Other features and effects of the present invention will become apparent from the following detailed description of the embodiments with reference to the accompanying drawings, in which:
FIG. 1 is a block diagram illustrating a first embodiment of the method for object recognition of moving images according to the present invention;
FIG. 2 is a flow chart of the first embodiment;
FIG. 3 is a schematic diagram illustrating the first embodiment for building a three-dimensional model;
FIG. 4 is a diagram illustrating the capturing of n key frames in an image sequence of a moving image according to the first embodiment;
FIG. 5 is a diagram illustrating the first embodiment intercepting a plurality of target images;
FIG. 6 is a block diagram illustrating a second embodiment of the method for interactive film creation with automatic capture of a target image according to the present invention;
FIG. 7 is a flow chart of the second embodiment;
fig. 8 is a diagram showing the setting of the content attribute by a setting layout according to the second embodiment; and
fig. 9 is a schematic diagram of the second embodiment for playing an interactive movie.
Detailed Description
Before the present invention is described in detail, it should be noted that in the following description, like elements are represented by like reference numerals.
Referring to fig. 1, a first embodiment of the object recognition method for dynamic images of the present invention uses a recognition system 1 as a tool. In this embodiment, the identification system 1 includes a first electronic device 11, and a storage medium 12 and an image capturing device 13 connected to the first electronic device 11. The first electronic device 11 can be a computer or a handheld second electronic device such as: notebook computers, tablet computers, touch computers, or mobile communication devices such as: a smart phone or a personal digital assistant device, and includes a communication module 111 and a processing module 112 electrically connected to the communication module 111. The storage medium 12 may be a hard disk installed on the first electronic device 11, or a removable portable disk connected to the first electronic device 11, a portable hard disk, or a cloud hard disk communicating with the first electronic device 11 through the communication module 111, or a storage device transmitting information through a wireless communication technology. The image capturing device 13 can be directly connected to the first electronic device 11, or can communicate with the first electronic device 11 through a wireless communication technology or a network.
Referring to fig. 2, and with reference to fig. 3, fig. 4 and fig. 5, the following description of the object recognition method for dynamic images according to the present invention with reference to the following steps is provided:
step 20: the processing module 112 of the first electronic device 11 starts the identification programming.
Step 21: the processing module 112 of the first electronic device 11 obtains a moving image 4 and an image 51. The format of the moving image 4 may be compiled by a web-based video format, such as flv, h.263, h.264, f4v, rm, wmv, m4v, mov, webm, etc., the format of the image 51 may be BMP, GIF, JPG/JPEG, PNG, TIF, etc., and the source of the image 51 may be the storage medium 12 connected to the first electronic device 11, or the storage medium 12 in the cloud, or directly capturing the frame in the moving image 4.
It is important that the source of the moving image 4 can be the aforementioned storage medium 12 connected to the first electronic device 11, or a cloud storage medium 12 such as youtube, youku, or a real-time image captured on-site by the image capturing device 13.
Step 22: the processing module 112 of the first electronic device 11 performs edge detection on the image 51 to determine a target object 511.
It should be noted that the edge detection technique of the image 51 can be implemented in a manner widely used in the art, and the details are not described since the extended details can be derived by a person skilled in the art according to the above description.
Step 23: the processing module 112 of the first electronic device 11 builds a stereo model 6 according to the image features of the target object 511.
It should be noted that the capturing of the image Features may use an image recognition and description algorithm (SURF), identify feature points with invariants of scaling and rotation in the image, calculate feature vectors of the feature points, compile the feature vectors into image feature codes, and obtain the image Features. The conversion of the target object 511 in the 2D image 51 into the 3D stereoscopic model 6 can be performed in a manner widely used in the art, and the expanded details are not described since the extended details can be deduced by a person skilled in the art according to the above description.
Step 24: the processing module 112 of the first electronic device 11 captures N key frames in the image sequence V1-Vx of the moving image 4, where N has an initial value of 1 and a maximum value N.
For example, if the video sequence V1 through Vx of the moving image 4 has 100 key pictures N, N is 100.
Step 25: the processing module 112 of the first electronic device 11 compares whether there is an object 41 with a similarity of 50% to 100% with the contour line in the key frame n according to the contour line of the three-dimensional model 6, if yes, go to step 26, and if no, go to step 29.
Step 26: the processing module 112 of the first electronic device 11 captures the image feature of the object 41 in the key frame n.
Step 27: the processing module 112 of the first electronic device 11 compares whether the image feature of the object 41 in the key frame n is consistent with any angle of the three-dimensional model 6, if yes, go to step 28, and if no, go to step 29.
Step 28: the processing module 112 of the first electronic device 11 captures the key frame n as a target image 52 (here, fig. 5 only illustrates the target image 52 by the object 41 of the key frame n), and knows the time point when the target image 52 appears in the image sequences V1-Vx of the moving image 4.
For example, when an object 41 corresponding to any angle in the stereoscopic model 6 appears in the 20 th key frame n-20 of the 100 key frames n, the 20 th key frame n is captured as the target image 52.
Step 29: the processing module 112 of the first electronic device 11 determines whether N is equal to the maximum value N, and if not, proceeds to step 30, and if so, proceeds to step 31.
Step 30: the processing module 112 of the first electronic device 11 sets n to n +1, and returns to step 25.
Step 31: the processing module 112 of the first electronic device 11 stops recognizing the programming.
According to the above, the present invention only needs to input the image 51 containing the target object 511 to find the corresponding and required object 41 in the arbitrary dynamic image 4, thereby, the present invention does not need machine learning time and database establishment. For example, if the present invention is applied to a traffic system, the monitor can be used to transmit the real-time image to the first electronic device 11 only by inputting an image of a license plate number, so as to quickly find out the vehicle having the license plate and the location of the vehicle. Or, only one person's head portrait needs to be input, the monitor can be used to transmit the real-time image to the first electronic device 11, so as to quickly find out the person who accords with the head portrait and the location of the person.
Referring to fig. 2, fig. 6, fig. 7, fig. 8, and fig. 9, a second embodiment of the interactive movie creation method for automatically capturing a target image according to the present invention is also provided, and the identification system 1 is used as a tool. In this embodiment, the identification system 1 further comprises a second electronic device 14 in communication with the first electronic device 11 via a network. The first electronic device 11 is a server in this embodiment. The second electronic device 14 can be a computer or a handheld second electronic device such as: notebook computers, tablet computers, touch computers, or mobile communication devices such as: the smart phone or the personal digital assistant device transmits at least one instruction through an application program, so that the first electronic device 11 executes the interactive film setup program according to the instruction and then step 31.
The following description is further directed to an interactive movie creation method for automatically capturing a target image by using an object recognition method of a moving image, in combination with the steps of a business application:
step 32: the processing module 112 of the first electronic device 11 starts interactive movie setup programming.
Step 33: the processing module 112 of the first electronic device 11 presents all of the target images 52.
Step 34: the processing module 112 of the first electronic device 11 presents a setting layout 7 pre-loaded with the moving image 4 and the target image 52 according to a trigger message from the second electronic device 14 for clicking any one of the target images 52.
Step 35: the processing module 112 of the first electronic device 11 sets the content attribute related to the target image 52 according to the condition input by the second electronic device 14 through the setting layout 7, wherein the content attribute at least includes a connection website, a name, a price, or the like.
Step 36: the processing module 112 of the first electronic device 11 combines the moving image 4 and the selected target image 52 into an interactive film.
Step 37: the processing module 112 of the first electronic device 11 issues an online video playback interface 8 through the communication module 111, and embeds the online video playback interface 8 into a web page or an application program through a network.
The online video/audio interface 8 includes a first frame 81 for browsing the moving image 4 and a second frame 82 for browsing the target image 52. The target image 52 is displayed in the second frame 82 in one of sequential display and synchronous display along the time axis of the moving picture 4 in the first frame 81.
Step 38: the processing module 112 of the first electronic device 11 stops the interactive movie setup programming.
It should be noted that the aforesaid interactive movie technology is disclosed in the patent disclosure of the invention of china No. CN103678448A, "interactive movie business publishing method and system", previously filed by the present inventor. However, the present invention can further improve the inconvenience of the chinese patent No. CN103678448A that the time point when the object appears and the image should be manually set.
From the above description, the advantages of the foregoing embodiments can be summarized as follows:
1. taking the first embodiment as an example, the present invention can be completely different from the conventional method, and can immediately identify the required object 41 for any unexpected moving image 4 without machine learning time and database establishment, so as to break the limitation that the prior art can only identify the 2D image 51, and the identification speed is fast (the identification speed can be 1/3-1/4 of the playing time of the moving image 4), accurate, and can be widely applied in the fields of daily life, business, etc.
2. Taking the second embodiment as an example, the present invention can create an interactive movie for a desired target image 52 and a dynamic image 4, and automatically search all target images 52 in the dynamic image 4 that match the image 51 and corresponding time points, so that a user does not need to completely browse the dynamic image 4, or to memorize or manually input the time points of the target images 52, thereby greatly shortening the time for creating the interactive movie, and improving convenience.
The above description is only an example of the present invention, and the scope of the present invention should not be limited thereby, and the invention is still within the scope of the present invention by simple equivalent changes and modifications made according to the claims and the contents of the specification.
Claims (13)
1. An object recognition method of dynamic images, which uses a recognition system as a tool, is characterized in that the object recognition method comprises the following steps:
step a: the identification system acquires a dynamic image and an image;
step b: the identification system detects the edge of the image to determine a target object;
step c: the identification system establishes a model according to the image characteristics of the target object;
step d: the identification system captures n key pictures in the image sequence of the dynamic image, wherein the initial value of n is 1;
step e: the identification system compares whether an object which is consistent with the model exists in the key picture, if so, the step f is carried out, and if not, the step g is carried out;
step f: the identification system captures the key picture as a target image; and
step g: the identification system makes n equal to n +1 and returns to step e.
2. The method for object recognition of moving images according to claim 1, wherein: the identification system of step f simultaneously obtains the time point of the target image appearing in the image sequence of the dynamic image.
3. The method for object recognition of moving images according to claim 1, wherein: the model established by the identification system in the step c is a three-dimensional model, and the identification system in the step e can compare whether an object which is consistent with any angle of the model exists in the key picture.
4. The method for object recognition of moving images according to claim 3, wherein: step e comprises
Step e-1: the identification system compares whether an object with the similarity of 50-100% with the contour line exists in the key picture according to the contour line of the three-dimensional model, if so, the step e-2 is carried out, if not, the step g is carried out,
step e-2: the recognition system captures the image feature of the object in the key picture,
step e-3: the identification system compares whether the image characteristics of the object in the key picture are consistent with any angle of the three-dimensional model, if so, the step f is carried out, and if not, the step g is carried out.
5. The method for object recognition of moving images according to claim 1, wherein: n has a maximum value, and step g comprises the steps g-1: the identification system judges whether n is equal to the maximum value, if yes, the step g-2 is carried out: the identification system stops identification, if not, the step g-3 is carried out: the identification system makes n equal to n +1 and returns to step e.
6. The method for object recognition of moving images according to claim 1, wherein: the dynamic image and the image come from any storage medium connected with the recognition system.
7. The method for object recognition of moving images according to claim 1, wherein: the identification system comprises an image capturing device for capturing the dynamic image, wherein the dynamic image is a real-time picture captured by the image capturing device on site.
8. An interactive film building method for automatically capturing a target image by using the object recognition method of moving images as claimed in claim 2, wherein the interactive film building method for automatically capturing the target image comprises the following steps:
step h after step f: the identification system combines the dynamic image and the selected target image into an interactive film;
step i: the identification system issues an online video-audio playing interface, and the online video-audio playing interface comprises a first frame for browsing the dynamic images and a second frame for browsing the target images.
9. The interactive film creation method for automatically capturing a target image as recited in claim 8, wherein: the identification system embeds the online video-audio playing interface into one of a webpage and an application program through a network.
10. The interactive film creation method for automatically capturing a target image as recited in claim 8, wherein: the target image is displayed in the second frame in one of sequential display and synchronous display along with the time axis of the dynamic image in the first frame.
11. The interactive film creation method for automatically capturing a target image as recited in claim 8, wherein: the interactive film establishing method for automatically intercepting the target image further comprises the following steps j-1 before and after step f: the identification system presents all target images, step j-2: the identification system presents a set layout preloaded with the dynamic image and the target image according to a trigger message of any one target image selected by a point, and the step j-3: the identification system sets the content attribute related to the target image according to the condition input by the setting layout, wherein the content attribute at least comprises a connection website.
12. The interactive film creation method for automatically capturing a target image as recited in claim 11, wherein: the content attributes also include name and price.
13. The interactive film creation method for automatically capturing a target image as recited in claim 8, wherein: the identification system comprises a first electronic device and a second electronic device connected with the first electronic device, wherein the second electronic device transmits at least one instruction through application programming, so that the first electronic device executes the steps according to the instruction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510818386.1A CN106778449B (en) | 2015-11-23 | 2015-11-23 | Object identification method of dynamic image and interactive film establishment method for automatically capturing target image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510818386.1A CN106778449B (en) | 2015-11-23 | 2015-11-23 | Object identification method of dynamic image and interactive film establishment method for automatically capturing target image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106778449A CN106778449A (en) | 2017-05-31 |
| CN106778449B true CN106778449B (en) | 2020-09-22 |
Family
ID=58963012
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510818386.1A Active CN106778449B (en) | 2015-11-23 | 2015-11-23 | Object identification method of dynamic image and interactive film establishment method for automatically capturing target image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106778449B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI662355B (en) * | 2018-03-12 | 2019-06-11 | 緯創資通股份有限公司 | Image background removing system, image background changing apparatus, image background removing method, and computer program product thereof |
| CN112788260A (en) * | 2019-11-11 | 2021-05-11 | 纽伦百仕科技股份有限公司 | Intelligent identification single-chip system combined with image sensor |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101639834A (en) * | 2008-08-01 | 2010-02-03 | 昆盈企业股份有限公司 | Automatic image identification digital photo frame, automatic image identification display system and automatic image identification display method |
| CN103093177A (en) * | 2011-10-28 | 2013-05-08 | 台湾色彩与影像科技股份有限公司 | Face identification, detection and monitoring method |
| CN103426048A (en) * | 2012-05-25 | 2013-12-04 | 陆意志 | Publishing method for digital publication and publishing system for digital publication |
| US20140029801A1 (en) * | 2011-04-12 | 2014-01-30 | National University Of Singapore | In-Video Product Annotation with Web Information Mining |
| US20140075274A1 (en) * | 2012-09-13 | 2014-03-13 | Yi-Chih Lu | Method for Publishing Composite Media Content and Publishing System to Perform the Method |
| CN104715254A (en) * | 2015-03-17 | 2015-06-17 | 东南大学 | Ordinary object recognizing method based on 2D and 3D SIFT feature fusion |
| CN105046219A (en) * | 2015-07-12 | 2015-11-11 | 上海微桥电子科技有限公司 | Face identification system |
-
2015
- 2015-11-23 CN CN201510818386.1A patent/CN106778449B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101639834A (en) * | 2008-08-01 | 2010-02-03 | 昆盈企业股份有限公司 | Automatic image identification digital photo frame, automatic image identification display system and automatic image identification display method |
| US20140029801A1 (en) * | 2011-04-12 | 2014-01-30 | National University Of Singapore | In-Video Product Annotation with Web Information Mining |
| CN103093177A (en) * | 2011-10-28 | 2013-05-08 | 台湾色彩与影像科技股份有限公司 | Face identification, detection and monitoring method |
| CN103426048A (en) * | 2012-05-25 | 2013-12-04 | 陆意志 | Publishing method for digital publication and publishing system for digital publication |
| US20140075274A1 (en) * | 2012-09-13 | 2014-03-13 | Yi-Chih Lu | Method for Publishing Composite Media Content and Publishing System to Perform the Method |
| CN104715254A (en) * | 2015-03-17 | 2015-06-17 | 东南大学 | Ordinary object recognizing method based on 2D and 3D SIFT feature fusion |
| CN105046219A (en) * | 2015-07-12 | 2015-11-11 | 上海微桥电子科技有限公司 | Face identification system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106778449A (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI582710B (en) | The method of recognizing the object of moving image and the interactive film establishment method of automatically intercepting target image | |
| US20210233319A1 (en) | Context-aware tagging for augmented reality environments | |
| CN106203242B (en) | Similar image identification method and equipment | |
| US9111255B2 (en) | Methods, apparatuses and computer program products for determining shared friends of individuals | |
| US20180107876A1 (en) | Method and system for providing augmented reality contents by using user editing image | |
| US10789474B2 (en) | System, method and apparatus for displaying information | |
| US9606975B2 (en) | Apparatus and method for automatically generating visual annotation based on visual language | |
| KR20130102549A (en) | Automatic media sharing via shutter click | |
| WO2020044097A1 (en) | Method and apparatus for implementing location-based service | |
| JP2021034003A (en) | Human object recognition method, apparatus, electronic device, storage medium, and program | |
| JP6046501B2 (en) | Feature point output device, feature point output program, feature point output method, search device, search program, and search method | |
| KR101782590B1 (en) | Method for Providing and Recommending Related Tag Using Image Analysis | |
| TW201631960A (en) | Display system, method, computer readable recording medium and computer program product for video stream on augmented reality | |
| WO2017107431A1 (en) | Method, device and electronic equipment for sorting images | |
| TW201327203A (en) | Methods and apparatus for progressive pattern matching in a mobile environment | |
| CN102541836B (en) | A kind of method and device adjusting format file layout | |
| CN111429338A (en) | Method, apparatus, device and computer-readable storage medium for processing video | |
| US20160105620A1 (en) | Methods, apparatus, and terminal devices of image processing | |
| KR101715708B1 (en) | Automated System for Providing Relation Related Tag Using Image Analysis and Method Using Same | |
| CN106778449B (en) | Object identification method of dynamic image and interactive film establishment method for automatically capturing target image | |
| TWM506428U (en) | Display system for video stream on augmented reality | |
| CN113869063A (en) | Data recommendation method and device, electronic equipment and storage medium | |
| CN111629267B (en) | Audio labeling method, device, equipment and computer readable storage medium | |
| Ou et al. | Machine learning-based object recognition technology for bird identification system | |
| CN106650727B (en) | Information display method and AR equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |