Remote Human Action Reproduction Using Prebuilt 3D Body Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing remote image reproduction methods face challenges in real-time reconstruction of three-dimensional human bodies, requiring significant computational resources and network bandwidth, leading to poor interaction experiences due to frame loss and freezing.
Innovation Solution
Pre-construct a realistic human body model during a preparation phase, generating action images using the model in the remote reproduction phase, reducing the need for real-time multi-angle image transmission and minimizing network resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If real-time three-dimensional human body reconstruction is performed using multi-angle images, then remote image reproduction is achieved, but computational resources and time are excessively consumed, resulting in poor reconstruction effect and failure to meet service requirements
Solution Approach 1:
The patent pre-constructs a three-dimensional human body model during an idle period before the remote reproduction task. This preliminary action allows the computationally intensive model construction to be performed when system resources are available, rather than in real-time during the actual remote reproduction task. The pre-constructed model is then stored and reused when needed, eliminating the time constraint during critical operations.
Solution Approach 2:
The patent separates the model construction process from the image reproduction process. The three-dimensional model construction is performed as a distinct preliminary step, while the actual remote reproduction only requires transmitting and displaying images based on the pre-built model. This segmentation allows resource-intensive operations to be decoupled from time-critical operations.
2Reliability
If multi-angle images are transmitted in real-time for remote reconstruction, then remote reproduction is achieved, but network resources are excessively occupied, leading to frame loss and freezing when network quality is poor
Solution Approach 1:
Instead of transmitting multiple original multi-angle images for reconstruction, the patent uses a pre-constructed three-dimensional model to generate or represent the remote image. This copying approach replaces data-intensive image transmission with model-based rendering, significantly reducing network bandwidth requirements while maintaining visual quality.
Solution Approach 2:
The three-dimensional model is constructed in advance during an idle period, so that during actual remote reproduction, only lightweight model data or pre-rendered views need to be transmitted. This preliminary preparation eliminates the need for real-time processing and transmission of heavy image data, ensuring stable performance even under poor network conditions.
Data Source
AI summary
This application provides a remote reproduction method, applied to a remote reproduction system, where the system includes a first terminal worn by a first user and a second terminal worn by a second user, and the first user and the second user are located in different geographical locations. According to the method the second terminal obtains an action image of the first user, where the action image of the first user includes a sequence of at least one action formed by a pre-constructed human body model of the first user, and then the second terminal reproduces, for the second user, the action image of the first user at a geographical location in which the second user is located.