WO2021075995A1 - Способ формирования поисковой выдачи в рекламном виджите - Google Patents
Способ формирования поисковой выдачи в рекламном виджите Download PDFInfo
- Publication number
- WO2021075995A1 WO2021075995A1 PCT/RU2019/000741 RU2019000741W WO2021075995A1 WO 2021075995 A1 WO2021075995 A1 WO 2021075995A1 RU 2019000741 W RU2019000741 W RU 2019000741W WO 2021075995 A1 WO2021075995 A1 WO 2021075995A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- objects
- features
- neural network
- search results
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000013528 artificial neural network Methods 0.000 claims abstract description 37
- 239000013598 vector Substances 0.000 claims abstract description 25
- 238000004458 analytical method Methods 0.000 claims abstract description 12
- 238000012549 training Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 8
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 abstract description 6
- 238000000605 extraction Methods 0.000 description 9
- 230000015572 biosynthetic process Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 210000002569 neuron Anatomy 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 241001025261 Neoraja caerulea Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0641—Shopping interfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0603—Catalogue ordering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/0895—Weakly supervised learning, e.g. semi-supervised or self-supervised learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
Definitions
- the present technical solution relates to the field of computing, in particular, to a method for generating search results in an advertising widget.
- the disadvantages of this solution are that it does not use a detector before using the neural network to calculate the vector representation.
- the use of the detector gives a significantly higher quality vector representation due to the clipping of the background and other objects that may be present in the image.
- the triplet generation method is based on using a random object as a negative example without further specifying how this random object is selected. If you just choose an arbitrary random object, then training will be extremely ineffective. Most triplets will be classified correctly in the early stages of learning and will not give any gain in the quality of the vector representation. At the same time, the effectiveness of training will be greatly slowed down.
- the technical problem to be solved by the claimed technical solution is the creation of a computer-implemented method of generating search results in an advertising widget, which is characterized in an independent claim. Additional embodiments of the present invention are presented in the dependent claims.
- the technical result consists in the reliability of object recognition from a context-media site for automatic search for relevant goods in electronic store catalogs.
- a computer-implemented method of generating search results in an advertising widget which consists in performing the steps at which, using at least one neural network (NN): - receive an image and a text description obtained from a context-media site;
- NN neural network
- vectors are calculated corresponding to the objects in the semantic space
- the selection of detected objects is carried out by bounding rectangles.
- the features of the original image that are not related to the selected object are suppressed by selecting the object along the contour.
- classifiers are formed at the training stage using a training sample, generating optimal classifiers.
- a neural network with the Mask R-CNN architecture is used to analyze the extracted features.
- a neural network trained on triplets is used to compute a vector in the semantic space.
- a neural network is additionally used to classify the image quality.
- relevant products are displayed to the user with the ability to go to a specific product page for purchases
- FIG. 1 illustrates a computer-implemented method of generating search results in an advertising widget
- FIG. 2 illustrates a diagram for analyzing content from a display site
- FIG. 3 illustrates an analysis diagram of a product catalog
- FIG. 4 illustrates the structure of the claimed solution
- FIG. 5 illustrates an example of a general arrangement of a computing device.
- An artificial neural network (hereinafter - ANN) is a computational or logical circuit built from homogeneous processing elements, which are simplified functional models of neurons.
- a neuron is a separate computational element of a network; each neuron is connected to the neurons of the previous and next layers of the network.
- each neuron is connected to the neurons of the previous and next layers of the network.
- the network can change its configuration (link weights, offset values, etc.).
- Artificial neural networks are an important tool for solving many applied problems. They have already made it possible to cope with a number of difficult problems and promise the creation of new inventions capable of solving problems that only man can do so far. Artificial neural networks, just like biological ones, are systems consisting of a huge number of functioning processors-neurons, each of which performs some small amount of work assigned to it, while having a large number of connections with the rest, which characterizes the power of network computing.
- a widget is a small graphic element or module that is inserted into a website or displayed on the desktop to display important and frequently updated information.
- Contextual media site is a system for placing contextual advertising and advertising that takes into account the interests of users on the pages of sites-participants of the partner network.
- the present invention is aimed at providing a computer-implemented method for generating search results in an advertising widget.
- the claimed computer-implemented method (100) is implemented as follows:
- step (101) an image and a text description obtained from the contextual media site are received.
- the obtained image of the investigated area is processed by detecting objects in the image, and features of objects in the image are distinguished.
- step (103) the selected features are analyzed, and on the basis of the analysis, the detected objects are extracted to separate them into classes.
- step (104) the features of the text description are distinguished. Using the signs of objects in the image and signs of a text description at the stage
- step (106) calculate vectors corresponding to objects in the semantic space.
- step (106) the obtained vector combination is used to search for relevant goods in electronic store catalogs.
- step (107) the search results are generated in the advertising widget.
- FIG. 2 shows a diagram of content analysis from a contextual media site, where at the first stage they carry out:
- the text associated with the image is analyzed (article test, image description): 1. Obtaining the text associated with the image (202) (for example, an image caption, text or article title);
- the result is obtained based on the results of the processes at the first and second stages:
- FIG. 3 shows a diagram of the analysis of the catalog of goods, where, at the first stage, the image in the product catalog is analyzed:
- the result is obtained based on the results of the processes at the first and second stages:
- a neural network for image feature extraction for example, a neural network with the architecture of ResNet, ResNeXt, MobileNet, etc. can be used, depending on the requirements for system performance and search quality.
- a network with the Mask R-CNN architecture can be used, which makes it possible to highlight the contours ("masks") of instances of different objects in the images, even if there are several such instances, they have different sizes and partially overlap.
- the LASER library can be used, which allows using texts in a large number of languages.
- the task of finding similar goods is reduced to the task of finding the nearest vectors in the metric space (kNN - k-nearest neighbors).
- the tasks of neural networks are to detect objects of interest to us in images and map each object into a certain vector in space while maintaining similarity. A similar approach is used in the face recognition problem.
- a specially collected and prepared dataset consisting of 2 million images is used for training.
- This set of images consists of: photos from websites, instagram and product catalogs. Images from product catalogs are matched with paired images from other sources. Pairs can be formed both from images of the same products, and similar ones. Most of the images have text descriptions.
- the resulting detector in the claimed solution was used to detect objects in all remaining images. Then, pairs of objects in these images were formed from pairs of images. A similarity score (rank) is associated with each pair.
- image processing begins with feature extraction, and this part of the neural network is used in all other stages. This creates additional learning difficulties. For the sake of simplicity, let's first consider the training of different warheads separately.
- the vector representation formation neural network is trained using triplets and triplet loss (FaceNet 2015, https://arxiv.org/abs/1503.03832). Triplets are generated automatically from the existing pairs of objects, taking into account the similarity assessment and the state of the neural network. The positive pair is taken from the database, and the negative pair is chosen randomly from the search results using the current version of the neural network.
- the input data for the neural network for the formation of a vector representation are the features of the original image reduced to the object's bounding rectangle (aligned feature maps), the object mask and the features of the textual description of the object.
- Training an image feature extraction neural network for such a variety of applications is not an easy task.
- the main difficulty is that learning to rank using triplets requires three times as much memory. Therefore, when teaching ranking, a lightweight version of the feature extraction neural network is used.
- training takes place sequentially for different head units. For each head, a certain number of steps are performed, then the head is changed to another and the process continues.
- User devices (401); 2. The web server of the contextual media site (402);
- the user device can be a personal computer, smartphone, TV or other devices with Internet access.
- the user device generates a request to display the widget, receives information about the content of the widget from the widget's web server (404), displays the widget, and interacts between the widget and the user.
- the user is redirected to the web server of the store's electronic catalog (403).
- the electronic store catalog also serves as a source of information for the indexing server (406), which periodically updates information about the products in the database (407). When new products are found, the index server analyzes them and calculates vector representations for them.
- the widget is formed on the side of the widget's web server. Several scenarios for the formation of the widget are possible. Let's consider the most typical ones.
- the widget is embedded in a contextual media site and displays product offers associated with the photos on that site.
- the search server (405) generates search results, which is stored in the database (407).
- the search results come from the database without any resource-intensive processing.
- the widget is embedded in a site or application and shows product offers associated with custom photos, which can be generated in real time.
- the formation of search results occurs online at the time the user device accesses the widget's web server.
- Web server widget accesses a search server which executes the process of FIG. 1.
- the steps (101) - (105) of the content analysis process may be transferred to the user device side.
- the widget's web server accepts only vector representations of objects instead of content.
- the widget is embedded in the video player and is activated when the video is paused or a special button is pressed. In this case, not one image can be analyzed, but a number of frames preceding this event.
- a source of text data can be used, for example, subtitles or converted into text audio. Processing can take place both online and offline. As in the previous case, a significant part of the computational load can be transferred to the user's device.
- FIG. 5 a general diagram of a computer device (500) will be presented that provides data processing necessary for the implementation of the claimed solution.
- the device (500) contains such components as: one or more processors (501), at least one memory (502), data storage means (503), input / output interfaces (504), I / O means ( 505), networking tools (506).
- the processor (501) of the device performs the basic computational operations necessary for the operation of the device (500) or the functionality of one or more of its components.
- the processor (501) executes the necessary computer readable instructions contained in the main memory (502).
- Memory (02), as a rule, is made in the form of RAM and contains the necessary program logic that provides the required functionality.
- the data storage medium (503) can be performed in the form of HDD, SSD disks, raid array, network storage, flash memory, optical information storage devices (CD, DVD, MD, Blue-Ray disks), etc.
- the means (503) allows performing long-term storage of various types of information, for example, the aforementioned files with user data sets, a database containing records of time intervals measured for each user, user identifiers, etc.
- Interfaces (504) are standard means for connecting and working with the server side, for example, USB, RS232, RJ45, LPT, COM, HDMI, PS / 2, Lightning, FireWire, etc.
- interfaces (504) depends on the specific implementation of the device (500), which can be a personal computer, mainframe, server cluster, thin client, smartphone, laptop, etc.
- a keyboard should be used.
- the hardware design of the keyboard can be any known: it can be either a built-in keyboard used on a laptop or netbook, or a stand-alone device connected to a desktop computer, server or other computer device.
- the connection can be either wired, in which the connecting cable of the keyboard is connected to the PS / 2 or USB port located on the system unit of the desktop computer, or wireless, in which the keyboard exchanges data via a wireless communication channel, for example, a radio channel, with base station, which, in turn, is directly connected to the system unit, for example, to one of the USB ports.
- I / O data can also include: joystick, display (touchscreen display), projector, touchpad, mouse, trackball, light pen, speakers, microphone, etc.
- Networking means (506) are selected from a device that provides network reception and transmission of data, for example, Ethernet card, WLAN / Wi-Fi module, Bluetooth module, BLE module, NFC module, IrDa, RFID module, GSM modem, etc.
- the means (505) the organization of data exchange via a wired or wireless data transmission channel is provided, for example, WAN, PAN, LAN, Intranet, Internet, WLAN, WMAN or GSM.
- the components of the device (500) are interconnected via a common data bus (510).
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Image Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
Claims
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/RU2019/000741 WO2021075995A1 (ru) | 2019-10-16 | 2019-10-16 | Способ формирования поисковой выдачи в рекламном виджите |
US17/627,610 US20220261856A1 (en) | 2019-10-16 | 2019-10-16 | Method for generating search results in an advertising widget |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/RU2019/000741 WO2021075995A1 (ru) | 2019-10-16 | 2019-10-16 | Способ формирования поисковой выдачи в рекламном виджите |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021075995A1 true WO2021075995A1 (ru) | 2021-04-22 |
Family
ID=75538569
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/RU2019/000741 WO2021075995A1 (ru) | 2019-10-16 | 2019-10-16 | Способ формирования поисковой выдачи в рекламном виджите |
Country Status (2)
Country | Link |
---|---|
US (1) | US20220261856A1 (ru) |
WO (1) | WO2021075995A1 (ru) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11989254B2 (en) * | 2020-09-10 | 2024-05-21 | Taboola.Com Ltd. | Semantic meaning association to components of digital content |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2473127C2 (ru) * | 2006-12-20 | 2013-01-20 | Майкрософт Корпорейшн | Интеграция рекламы и расширяемые темы для операционных систем |
US8781887B2 (en) * | 2007-11-26 | 2014-07-15 | Raymond Ying Ho Law | Method and system for out-of-home proximity marketing and for delivering awarness information of general interest |
WO2016037278A1 (en) * | 2014-09-10 | 2016-03-17 | Sysomos L.P. | Systems and methods for continuous analysis and procurement of advertisement campaigns |
RU2595597C2 (ru) * | 2011-09-29 | 2016-08-27 | Амазон Текнолоджис, Инк. | Электронная торговая площадка размещаемых образов услуг |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5121599B2 (ja) * | 2008-06-30 | 2013-01-16 | キヤノン株式会社 | 画像処理装置、画像処理方法およびそのプログラムならびに記憶媒体 |
WO2017153354A1 (de) * | 2016-03-07 | 2017-09-14 | SensoMotoric Instruments Gesellschaft für innovative Sensorik mbH | Verfahren und vorrichtung zum bewerten von blickabbildungen |
KR20190117584A (ko) * | 2017-02-09 | 2019-10-16 | 페인티드 도그, 인크. | 스트리밍 비디오 내의 객체를 검출하고, 필터링하고 식별하기 위한 방법 및 장치 |
CN108038880B (zh) * | 2017-12-20 | 2019-12-13 | 百度在线网络技术(北京)有限公司 | 用于处理图像的方法和装置 |
US11003856B2 (en) * | 2018-02-22 | 2021-05-11 | Google Llc | Processing text using neural networks |
US10902051B2 (en) * | 2018-04-16 | 2021-01-26 | Microsoft Technology Licensing, Llc | Product identification in image with multiple products |
US11244205B2 (en) * | 2019-03-29 | 2022-02-08 | Microsoft Technology Licensing, Llc | Generating multi modal image representation for an image |
-
2019
- 2019-10-16 US US17/627,610 patent/US20220261856A1/en not_active Abandoned
- 2019-10-16 WO PCT/RU2019/000741 patent/WO2021075995A1/ru active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2473127C2 (ru) * | 2006-12-20 | 2013-01-20 | Майкрософт Корпорейшн | Интеграция рекламы и расширяемые темы для операционных систем |
US8781887B2 (en) * | 2007-11-26 | 2014-07-15 | Raymond Ying Ho Law | Method and system for out-of-home proximity marketing and for delivering awarness information of general interest |
RU2595597C2 (ru) * | 2011-09-29 | 2016-08-27 | Амазон Текнолоджис, Инк. | Электронная торговая площадка размещаемых образов услуг |
WO2016037278A1 (en) * | 2014-09-10 | 2016-03-17 | Sysomos L.P. | Systems and methods for continuous analysis and procurement of advertisement campaigns |
Also Published As
Publication number | Publication date |
---|---|
US20220261856A1 (en) | 2022-08-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Katsurai et al. | Image sentiment analysis using latent correlations among visual, textual, and sentiment views | |
US11514235B2 (en) | Information extraction from open-ended schema-less tables | |
JP6397144B2 (ja) | 画像からの事業発見 | |
Yuan et al. | Sentribute: image sentiment analysis from a mid-level perspective | |
US8649602B2 (en) | Systems and methods for tagging photos | |
US11182806B1 (en) | Consumer insights analysis by identifying a similarity in public sentiments for a pair of entities | |
US10685183B1 (en) | Consumer insights analysis using word embeddings | |
US20160034512A1 (en) | Context-based metadata generation and automatic annotation of electronic media in a computer network | |
Chen et al. | Discovering informative social subgraphs and predicting pairwise relationships from group photos | |
WO2018184518A1 (zh) | 微博数据处理方法、装置、计算机设备及存储介质 | |
US20160242699A1 (en) | System and method for evaluating a cognitive load on a user corresponding to a stimulus | |
US9817904B2 (en) | Method and system for generating augmented product specifications | |
CN109271624A (zh) | 一种目标词确定方法、装置及存储介质 | |
US11030539B1 (en) | Consumer insights analysis using word embeddings | |
US20200097759A1 (en) | Table Header Detection Using Global Machine Learning Features from Orthogonal Rows and Columns | |
WO2021075995A1 (ru) | Способ формирования поисковой выдачи в рекламном виджите | |
Huang et al. | Analyzing group-level emotion with global alignment kernel based approach | |
Dogariu et al. | A Textual Filtering of HOG-Based Hierarchical Clustering of Lifelog Data. | |
Maynard et al. | Entity-based opinion mining from text and multimedia | |
US20170293863A1 (en) | Data analysis system, and control method, program, and recording medium therefor | |
JP6178480B1 (ja) | データ分析システム、その制御方法、プログラム、及び、記録媒体 | |
Shipman et al. | Towards a distributed digital library for sign language content | |
Su et al. | Cross-modality based celebrity face naming for news image collections | |
US11042520B2 (en) | Computer system | |
CN113704623A (zh) | 一种数据推荐方法、装置、设备及存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19949330 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19949330 Country of ref document: EP Kind code of ref document: A1 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19949330 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 20.10.2022) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19949330 Country of ref document: EP Kind code of ref document: A1 |