Method, storage medium and apparatus for converting 2d photo album to 3D model

A picture and 3D technology applied in the computer field to achieve the effect of enhancing entertainment and fun and improving user experience

Active Publication Date: 2019-01-11
SAMSUNG ELECTRONICS CHINA R&D CENT +1
View PDF2 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In view of this, the present invention provides a method, storage medium and device for converting a 2D image collection to a 3D model, to solve the problem of how to construct a 3D model based on a 2D image collection

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method, storage medium and apparatus for converting 2d photo album to 3D model
  • Method, storage medium and apparatus for converting 2d photo album to 3D model
  • Method, storage medium and apparatus for converting 2d photo album to 3D model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0110] The user shoots a 2D video of a child playing with a mobile phone. Based on the above-mentioned method of the present invention, the video can be converted into a section of AR content, and the AR content can be directly watched to experience the feeling of "being there". The specific operation is implemented as follows:

[0111] Step 201: analyze the 2D video; open the video, obtain the static frame of the video, analyze the static frame, find out the key frame, and save it, such as Figure 6 shown.

[0112] Step 202: Identify and extract the child in the key frame and related feature information through deep learning algorithm and computer vision algorithm, such as Figure 7 shown.

[0113] When extracting, you can choose not to set the target object, and use the default target. The default target is: people, cars, chairs, cups, birds, cows, cats, dogs, sheep, etc. You can also choose to set a specific extraction target, such as extracting only human target objects...

Embodiment 2

[0120] Sometimes due to some reasons, the user cannot go to the scene to watch the auto show, and can only watch the ordinary video of the car taken by a friend on the auto show. Based on the method of the present invention, the video can be converted into a piece of AR content, so that the user can watch the car like the experience on the spot.

[0121] Step 301: analyze the 2D video; open the video, obtain the static frame of the video, analyze the static frame, find out the key frame, and save it.

[0122] Step 302: Identify and extract the car and related feature information in the key frame through deep learning algorithm and computer vision algorithm.

[0123] When extracting, you can choose not to set the target object, and use the default target. The default target is: people, cars, chairs, cups, birds, cows, cats, dogs, sheep, etc. You can also choose to set a specific extraction target as a car, and only extract the target object and related features of the car. Al...

Embodiment 3

[0130] Users often watch some 2D performance videos. Through the method of the present invention, this video can be converted into a piece of AR content, so that users or others can experience the feeling of watching on the spot and experience the feeling of "being on the scene".

[0131] Step 401: analyze the 2D video; open the video, obtain the static frame of the video, analyze the static frame, find out the key frame, and save it.

[0132] Step 402: Identify and extract the stage and related feature information in the key frame through deep learning algorithm and computer vision algorithm.

[0133] When extracting, you can choose not to set the target object, and use the default target. The default target is: people, cars, chairs, cups, birds, cows, cats, dogs, sheep, etc. You can also choose to set a specific extraction target as the stage, and only extract the target object and related features of the stage. Alternatively, manually selected target objects and related f...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a method, a storage medium and a device for converting a 2D picture set to a 3D model. The method comprises the following steps: step 11, identifying the category of each objectcontained in each picture in a 2D picture set and the outline information of each object by a deep learning algorithm; step 12, extracting that detail characteristic information of each object in each picture through a computer vision algorithm, wherein the detail characteristic information comprise at least texture information, color characteristic information and shape characteristic information of each object; step 13, matching the 3D model based on the category, the shape characteristic information and the contour information of each object, and matching the 3D model successfully as the 3D model of each object; step 14, mapping the texture information and color feature information of each object onto the 3D model of each object. Based on the method of the invention, a realistic 3D model is constructed to overcome the shortcomings of generating 3D model and 3D video based on parallax, to improve the user experience of 3D video or 3D model, and to enhance entertainment and interest.

Description

technical field [0001] The invention relates to the field of computers, in particular to a method, a storage medium and a device for converting a 2D picture collection into a 3D model. Background technique [0002] At present, 2D pictures or videos can be converted into 3D models or 3D videos based on the principle of parallax, but the essence is to generate two 2D pictures for the left and right eyes differently, without generating a substantial 3D model. Because parallax is an illusion, users will feel uncomfortable, distorted, and easily fatigued when watching 3D models or 3D videos generated based on parallax, and the user experience is poor, which affects entertainment and interest. Contents of the invention [0003] In view of this, the present invention provides a method, storage medium and device for converting a 2D image collection to a 3D model, so as to solve the problem of how to construct a 3D model based on a 2D image collection. [0004] The present inventi...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06T3/00
CPCG06T3/005G06T2200/04G06N3/08G06N20/00G06N3/044G06N3/045G06T7/70G06T15/04G06T17/00G06T19/006G06T19/20G06T2200/08G06T2207/20084G06T2207/30244G06T2219/2012G06V20/46G06F18/23G06F18/24G06F18/213
Inventor 王磊
Owner SAMSUNG ELECTRONICS CHINA R&D CENT
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap