FIELD: physics.
SUBSTANCE: invention relates to information technology, specifically to means of creating 3D model of a human. System for generating video with reconstructed photorealistic 3D human model comprises the following functionally interconnected: a video preprocessing module configured for: a) sampling frames for further processing from an input video clip with a person rotating in front of the camera, with an indication of the reference frame in the selected frames; b) detecting a human; c) human image segmentation; d) determining on the segmented image the location of the reference points at which the joints of the human body with the body posture and shape are located; e) obtaining parameters of human body posture, human body shape and camera parameters based on data including a segmented image of a person and found locations of reference points; f) predicting 3D human mesh based on body posture parameters, body shape and camera parameters.
EFFECT: high quality and speed of creating photorealistic 3D model of a person.
7 cl, 14 dwg, 4 tbl
Title | Year | Author | Number |
---|---|---|---|
METHOD OF CREATING FULL-LENGTH ANIMATED AVATAR OF PERSON FROM ONE IMAGE OF PERSON, COMPUTING DEVICE AND MACHINE-READABLE MEDIUM FOR IMPLEMENTATION THEREOF | 2023 |
|
RU2813485C1 |
MODELLING PEOPLE'S CLOTHING BASED ON MULTIPLE POINTS | 2021 |
|
RU2776825C1 |
METHOD FOR VISUALIZING A 3D PORTRAIT OF A PERSON WITH ALTERED LIGHTING AND A COMPUTING DEVICE FOR IT | 2021 |
|
RU2757563C1 |
TEXTURED NEURAL AVATARS | 2019 |
|
RU2713695C1 |
NEURAL DOT GRAPHIC | 2019 |
|
RU2729166C1 |
RAPID TWO-LAYER NEURAL NETWORK SYNTHESIS OF REALISTIC IMAGES OF A NEURAL AVATAR BASED ON A SINGLE IMAGE | 2020 |
|
RU2764144C1 |
METHOD FOR RECONSTRUCTION OF 3D-MODEL OF OBJECT | 2020 |
|
RU2779271C2 |
NEURAL-NETWORK RENDERING OF THREE-DIMENSIONAL HUMAN AVATARS | 2021 |
|
RU2775825C1 |
METHOD AND SYSTEM FOR REMOTE CLOTHING SELECTION | 2020 |
|
RU2805003C2 |
REPEATED SYNTHESIS OF IMAGE USING DIRECT DEFORMATION OF IMAGE, PASS DISCRIMINATOR AND COORDINATE-BASED REMODELLING | 2019 |
|
RU2726160C1 |
Authors
Dates
2025-02-04—Published
2024-08-02—Filed