CN117156295A - Cloud shooting system based on deep neural network and intelligent template generation method - Google Patents

Cloud shooting system based on deep neural network and intelligent template generation method Download PDF

Info

Publication number
CN117156295A
CN117156295A CN202311113824.5A CN202311113824A CN117156295A CN 117156295 A CN117156295 A CN 117156295A CN 202311113824 A CN202311113824 A CN 202311113824A CN 117156295 A CN117156295 A CN 117156295A
Authority
CN
China
Prior art keywords
template
user
neural network
intelligent
deep neural
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
CN202311113824.5A
Other languages
Chinese (zh)
Inventor
李鑫
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.)
Shenzhen Huohuohuotu Intelligent Technology Co ltd
Original Assignee
Shenzhen Huohuohuotu Intelligent Technology Co ltd
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 Shenzhen Huohuohuotu Intelligent Technology Co ltd filed Critical Shenzhen Huohuohuotu Intelligent Technology Co ltd
Priority to CN202311113824.5A priority Critical patent/CN117156295A/en
Publication of CN117156295A publication Critical patent/CN117156295A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
    • G06V10/449—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters
    • G06V10/451—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters with interaction between the filter responses, e.g. cortical complex cells
    • G06V10/454—Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/66—Remote control of cameras or camera parts, e.g. by remote control devices
    • H04N23/661—Transmitting camera control signals through networks, e.g. control via the Internet
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/95—Computational photography systems, e.g. light-field imaging systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Databases & Information Systems (AREA)
  • Computer Graphics (AREA)
  • Geometry (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Processing Or Creating Images (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a cloud shooting system based on a deep neural network and an intelligent template generation method, which comprises the following steps: the intelligent terminal D1 is used for the user to shoot by himself; the cloud server C1 establishes long connection with the intelligent terminal D1, and the cloud server C1 is preset with a plurality of deep neural network models; the template management application A1 is used for the merchant to edit and generate an intelligent template in a self-defined way; the user application A2 is used for generating operation instructions and managing personal data for the user. Compared with the prior art, the method and the device can automatically generate the personalized intelligent template according to the materials provided by different users, reduce potential safety hazards of data, realize cross-terminal multiplexing and better meet application requirements.

Description

Cloud shooting system based on deep neural network and intelligent template generation method
Technical Field
The invention relates to a self-timer terminal cloud shooting system, in particular to a cloud shooting system based on a deep neural network and an intelligent template generation method.
Background
Currently, a self-timer terminal on the market is generally shipped from a factory to preset a template, and the factory uniformly maintains a photo template, and related contents are disclosed in a Chinese patent publication with publication number of CN112995629B and name of "an intelligent self-timer house implementation method based on a holographic technique", wherein:
the technical scheme is that the intelligent self-shooting library implementation method based on the holographic shooting technology comprises a holographic projection end and a shooting end, wherein the holographic projection end comprises a curtain, a projection module and a background module, a user can store background pictures into a storage unit according to the needs of the user, then the background pictures are projected into a space in front of the curtain through the projection module, the background pictures can be continuously replaced, a scene in the self-shooting library is not required to be removed and replaced, only the background pictures are required to be replaced for holographic projection, and the time required for replacing the scene is saved;
firstly, agents and service stores cannot customize templates according to own requirements; secondly, user data is reserved in the equipment, and privacy safety hidden danger exists; in addition, the user data cannot be multiplexed across terminals, and personalized application requirements cannot be met.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides the cloud shooting system based on the deep neural network and the intelligent template generation method, which can automatically generate personalized intelligent templates according to materials provided by different users, reduce potential safety hazards of data and realize cross-terminal multiplexing.
In order to solve the technical problems, the invention adopts the following technical scheme.
A cloud shooting system based on a deep neural network comprises: the intelligent terminal D1 is used for the user to shoot by himself; the cloud server C1 establishes long connection with the intelligent terminal D1, and the cloud server C1 is preset with a plurality of deep neural network models; the template management application A1 is used for the merchant to edit and generate an intelligent template in a self-defined way; the user application A2 is used for generating operation instructions and managing personal data for the user.
Preferably, the deep neural network model comprises object recognition, image segmentation, image classification, 3D reconstruction and/or image enhancement.
The intelligent template generation method based on the deep neural network is realized based on a cloud shooting system, wherein the cloud shooting system comprises an intelligent terminal D1, a cloud server C1, a template management application A1 and a user application A2, and the method comprises the following steps: step S1, a user logs in the cloud shooting system through the intelligent terminal D1; step S2, a user selects a shooting template provided by the template management application A1 through the intelligent terminal D1; step S3, a user controls a preset image capturing unit in the intelligent terminal D1, and photographs are captured according to requirements of the photographing template; step S4, submitting the photo to the cloud server C1 through the intelligent terminal D1 after shooting is completed; step S5, the cloud server C1 runs a cloud virtual shooting flow according to the shooting template to generate a sheet; step S6, after the virtual shooting is completed, the cloud server C1 informs a user by using a preset notification service; and S7, the user previews the film through the user application A2 or prints the photo through the intelligent terminal D1.
Preferably, in step S1, the user logs in the cloud shooting system by scanning a two-dimensional code on the intelligent terminal D1.
Preferably, in the step S4, if the intelligent terminal D1 does not operate within 3 seconds to 5 seconds, the login state is automatically exited.
Preferably, in the step S2, the merchant administrator uploads the material to the cloud server C1 through the template management application A1, and the cloud server C1 automatically generates the shooting template by using a preset deep neural network model.
Preferably, in the step S3, the user selects a mapping relationship between the user data and the template role through the intelligent terminal D1, and then submits a virtual shooting request.
Preferably, in step S7, the user scans the two-dimensional code for obtaining the slice through the intelligent terminal D1.
Preferably, in the step S4, after the shooting is completed, the photo data of the user is stored in an encrypted form in the cloud server C1.
Preferably, in the step S5, the virtual shooting process performed by the cloud server C1 uses only the picture feature data, and does not use the original photograph.
In the cloud shooting system based on the deep neural network disclosed by the invention, the intelligent terminal D1 can provide self-service shooting service for a user, the cloud server C1 is a cloud virtual shooting system which is in long connection with the intelligent terminal D1 and comprises a plurality of deep learning models, the template management application A1 is used as a generation tool for merchant custom editing intelligent templates, and the user can freely operate and manage personal data and application programs of the terminal through the user application A2. In the execution process, the system firstly collects material pictures provided by merchants, then uses a pre-trained deep neural network model to extract picture features, inputs the extracted feature vectors into another trained deep neural network model, then obtains template parameters, and finally generates a customizable intelligent template according to the obtained template parameters. Compared with the prior art, the method and the device can automatically generate the personalized intelligent template according to the materials provided by different users, reduce potential safety hazards of data, realize cross-terminal multiplexing and better meet application requirements.
Drawings
Fig. 1 is a schematic diagram of a cloud shooting system according to the present invention;
FIG. 2 is a flow chart of a method for generating a smart template according to a first embodiment of the present invention;
fig. 3 is a flowchart of a method for generating a smart template according to a second embodiment of the present invention.
Detailed Description
The invention is described in more detail below with reference to the drawings and examples.
The invention discloses a cloud shooting system based on a deep neural network, please refer to fig. 1, which comprises:
the intelligent terminal D1 is used for the user to shoot by himself;
the cloud server C1 establishes long connection with the intelligent terminal D1, and the cloud server C1 is preset with a plurality of deep neural network models;
the template management application A1 is used for the merchant to edit and generate an intelligent template in a self-defined way;
the user application A2 is used for generating operation instructions and managing personal data for the user.
In the above system, the intelligent terminal D1 may provide self-help shooting service for a user, the cloud server C1 is a cloud virtual shooting system that is long-connected with the intelligent terminal D1, and includes multiple deep learning models, the template management application A1 is used as a generating tool for a merchant to edit an intelligent template in a custom manner, and the user freely operates and manages personal data and an application program of the terminal through the user application A2. In the execution process, the system firstly collects material pictures provided by merchants, then uses a pre-trained deep neural network model to extract picture features, inputs the extracted feature vectors into another trained deep neural network model, then obtains template parameters, and finally generates a customizable intelligent template according to the obtained template parameters. Compared with the prior art, the method and the device can automatically generate the personalized intelligent template according to the materials provided by different users, reduce potential safety hazards of data, realize cross-terminal multiplexing and better meet application requirements.
As a preferred approach, the deep neural network model includes object recognition, image segmentation, image classification, 3D reconstruction, and/or image enhancement. In addition, model optimization and deployment are needed after model training is completed, so that the efficiency of online template generation is ensured.
On the basis, the invention also relates to an intelligent template generation method based on the deep neural network, referring to fig. 1, the method is realized based on a cloud shooting system, the cloud shooting system comprises an intelligent terminal D1, a cloud server C1, a template management application A1 and a user application A2, and the method comprises the following steps:
step S1, a user logs in the cloud shooting system through the intelligent terminal D1;
step S2, a user selects a shooting template provided by the template management application A1 through the intelligent terminal D1;
step S3, a user controls a preset image capturing unit in the intelligent terminal D1, and photographs are captured according to requirements of the photographing template;
step S4, submitting the photo to the cloud server C1 through the intelligent terminal D1 after shooting is completed;
step S5, the cloud server C1 runs a cloud virtual shooting flow according to the shooting template to generate a sheet;
step S6, after the virtual shooting is completed, the cloud server C1 informs a user by using a preset notification service;
and S7, the user previews the film through the user application A2 or prints the photo through the intelligent terminal D1.
In the step S1 of the present invention, a user logs in the cloud shooting system by scanning the two-dimensional code on the intelligent terminal D1.
In order to protect the privacy of the user, in the step S4 of the present invention, if the intelligent terminal D1 is not operated within 3 seconds to 5 seconds, the login state is automatically exited.
Further, in the step S4, after the photographing is completed, the photograph data of the user is stored in an encrypted form in the cloud server C1. In the step S5, the virtual shooting process executed by the cloud server C1 uses only the picture feature data and does not use the original photo, thereby realizing cross-terminal multiplexing.
In a preferred manner, in the step S2, the merchant administrator uploads the material to the cloud server C1 through the template management application A1, and the cloud server C1 automatically generates the shooting template by using a preset deep neural network model.
In the step S3 of the present invention, the user selects the mapping relationship between the user data and the template role through the intelligent terminal D1, and then submits the virtual shooting request.
Specifically, in the step S7, the user scans the two-dimensional code for obtaining the slice through the intelligent terminal D1.
The intelligent template generation method based on the deep neural network can be referred to the following embodiments in practical application:
example 1
Referring to fig. 2, the present embodiment includes the following steps:
step 1.1, a user scans a two-dimensional code on an intelligent terminal D1 and logs in a cloud shooting system;
step 1.2, a user selects a required cloud shooting template on an intelligent terminal D1;
step 1.3, a user takes a picture of a person role corresponding to the template by using a camera unit in the intelligent terminal D1;
step 1.4, the intelligent terminal D1 submits the photo to the cloud server C1 for virtual shooting, and during the period, if the intelligent terminal D1 is not operated for 3-5 seconds, the system login state can be automatically exited;
step 1.5, after the cloud server C1 completes the shooting task, informing a user through a notification service;
step 1.6, the user previews the sheeting in the user application A2.
In this embodiment, the intelligent template of the intelligent terminal D1 is automatically generated by using the deep neural network model by uploading the material to the cloud server C1 by the merchant administrator through the template management application A1.
Example two
Referring to fig. 3, the present embodiment includes the following steps:
step 2.1, a user scans a two-dimensional code on the intelligent terminal D1, logs in a cloud shooting system and loads user data;
step 2.2, a user selects a required cloud shooting template through the intelligent terminal D1;
step 2.3, the user selects the mapping relation between the user data and the template role through the intelligent terminal D1, submits a virtual shooting request, and automatically exits from a system login state if the intelligent terminal D1 is not operated for 3-5 seconds during the period;
step 2.4, after receiving the request, the cloud server C1 operates a cloud virtual shooting flow to finish cloud shooting;
step 2.5, the cloud server C1 informs a user through a notification service, and the user previews and prints photos through the user application A2;
step 2.6, the user scans the two-dimensional code of the user application A2 by using a scanning unit in the intelligent terminal D1;
and 2.7, the printing unit of the intelligent terminal D1 receives the printing instruction and prints the photo.
In this embodiment, the user may directly use the photo data stored in the cloud end, so that the step of repeated shooting is omitted. Meanwhile, the photo data of the user is stored in the cloud server C1 in an encrypted form, so that certain privacy protection is provided. In addition, the virtual shooting process in the cloud server C1 also uses only the picture feature data, and does not use the original picture file, thereby realizing cross-terminal multiplexing.
According to the cloud shooting system and the intelligent template generation method based on the deep neural network, provided by the invention, merchant custom templates are supported, the individuation degree of the templates is effectively improved, meanwhile, user data is stored in the cloud end in an encrypted manner, the privacy leakage risk is reduced, in addition, the cross-terminal multiplexing of the user data is supported, and the operation flow is simplified.
The above embodiments are only preferred embodiments of the present invention, and are not intended to limit the present invention, and modifications, equivalent substitutions or improvements made within the technical scope of the present invention should be included in the scope of the present invention.

Claims (10)

1. Cloud shooting system based on degree of depth neural network, characterized by including:
the intelligent terminal D1 is used for the user to shoot by himself;
the cloud server C1 establishes long connection with the intelligent terminal D1, and the cloud server C1 is preset with a plurality of deep neural network models;
the template management application A1 is used for the merchant to edit and generate an intelligent template in a self-defined way;
the user application A2 is used for generating operation instructions and managing personal data for the user.
2. The deep neural network-based cloud capture system of claim 1, wherein the deep neural network model comprises object recognition, image segmentation, image classification, 3D reconstruction, and/or image enhancement.
3. The intelligent template generation method based on the deep neural network is characterized by being realized based on a cloud shooting system, wherein the cloud shooting system comprises an intelligent terminal D1, a cloud server C1, a template management application A1 and a user application A2, and the method comprises the following steps:
step S1, a user logs in the cloud shooting system through the intelligent terminal D1;
step S2, a user selects a shooting template provided by the template management application A1 through the intelligent terminal D1;
step S3, a user controls a preset image capturing unit in the intelligent terminal D1, and photographs are captured according to requirements of the photographing template;
step S4, submitting the photo to the cloud server C1 through the intelligent terminal D1 after shooting is completed;
step S5, the cloud server C1 runs a cloud virtual shooting flow according to the shooting template to generate a sheet;
step S6, after the virtual shooting is completed, the cloud server C1 informs a user by using a preset notification service;
and S7, the user previews the film through the user application A2 or prints the photo through the intelligent terminal D1.
4. The intelligent template generating method based on the deep neural network according to claim 3, wherein in the step S1, a user logs in the cloud shooting system by scanning a two-dimensional code on the intelligent terminal D1.
5. The intelligent template generating method based on the deep neural network according to claim 3, wherein in the step S4, if the intelligent terminal D1 does not operate within 3 seconds to 5 seconds, the login state is automatically exited.
6. The intelligent template generating method based on the deep neural network according to claim 3, wherein in the step S2, a merchant administrator uploads the material to the cloud server C1 through the template management application A1, and the cloud server C1 automatically generates the shooting template by using a preset deep neural network model.
7. The intelligent template generation method based on the deep neural network according to claim 3, wherein in the step S3, the user selects a mapping relationship between user data and template roles through the intelligent terminal D1, and then submits a virtual shooting request.
8. The intelligent template generation method based on the deep neural network according to claim 3, wherein in the step S7, the user scans the slice-taking two-dimensional code through the intelligent terminal D1 and obtains slices.
9. The intelligent template generating method based on the deep neural network according to claim 3, wherein in the step S4, after the shooting is completed, the photo data of the user is stored in an encrypted form in the cloud server C1.
10. The intelligent template generating method based on the deep neural network according to claim 3, wherein in the step S5, the virtual photographing process performed by the cloud server C1 uses only picture feature data and does not use an original photograph.
CN202311113824.5A 2023-08-31 2023-08-31 Cloud shooting system based on deep neural network and intelligent template generation method Pending CN117156295A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311113824.5A CN117156295A (en) 2023-08-31 2023-08-31 Cloud shooting system based on deep neural network and intelligent template generation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311113824.5A CN117156295A (en) 2023-08-31 2023-08-31 Cloud shooting system based on deep neural network and intelligent template generation method

Publications (1)

Publication Number Publication Date
CN117156295A true CN117156295A (en) 2023-12-01

Family

ID=88898220

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311113824.5A Pending CN117156295A (en) 2023-08-31 2023-08-31 Cloud shooting system based on deep neural network and intelligent template generation method

Country Status (1)

Country Link
CN (1) CN117156295A (en)

Similar Documents

Publication Publication Date Title
US10067500B2 (en) Generating of 3D-printed custom wearables
CN100438580C (en) Method and apparatus for converting a photo to a caricature image
JP2017531950A (en) Method and apparatus for constructing a shooting template database and providing shooting recommendation information
CN112702521B (en) Image shooting method and device, electronic equipment and computer readable storage medium
CN109064545A (en) The method and device that a kind of pair of house carries out data acquisition and model generates
JP2004030668A (en) Portrait image correcting method
EP4013034A1 (en) Image capturing method and apparatus, and computer device and storage medium
KR101711684B1 (en) 3d avatars output device and method
CN112839175B (en) Method and device for acquiring image by cloud mobile phone, computer equipment and medium
MX2011009714A (en) Method and apparatus for video authentication of user.
CN110493512B (en) Photographic composition method, photographic composition device, photographic equipment, electronic device and storage medium
CN112598580A (en) Method and device for improving definition of portrait photo
CN112532911A (en) Image data processing method, device, equipment and storage medium
CN105578020B (en) Self-timer system and method
CN117156295A (en) Cloud shooting system based on deep neural network and intelligent template generation method
CN111294502A (en) Photographing method, device with photographing function, equipment and storage medium
CN108513034A (en) Method, electronic device and the computer readable storage medium of long-range shooting picture
CN111185903A (en) Method and device for controlling mechanical arm to draw portrait and robot system
KR102475520B1 (en) 3D modeling conversion method and system for realization of photorealistic based metaverse
CN112734657B (en) Cloud group photo method and device based on artificial intelligence and three-dimensional model and storage medium
CN115519792A (en) Simulation doll forming method, system, computer equipment and storage medium
CN113938597A (en) Face recognition method and device, computer equipment and storage medium
CN113393545A (en) Image animation processing method and device, intelligent device and storage medium
WO2019205566A1 (en) Method and device for displaying image
CN113784039B (en) Head portrait processing method, head portrait processing device, electronic equipment and computer readable storage medium

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