Digital human augmentation rendering method and system based on low-resolution video
By constructing a digital human rendering feature field interaction model, the problem of insufficient processing of inter-frame dynamic correlation in low-resolution digital human videos was solved, generating high-definition digital human videos with clear morphological details and smooth movements, thus improving the user experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- METADIGITAL SICHUAN TECHNOLOGY CO LTD
- Filing Date
- 2025-10-29
- Publication Date
- 2026-07-14
AI Technical Summary
Existing video enhancement technologies cannot effectively handle the dynamic inter-frame relationships in low-resolution digital human videos, resulting in insufficient smoothness of movements and consistency of morphological details, which cannot meet the needs of high-quality digital human video display.
A digital human rendering feature field interaction model is constructed. The morphological and motion features of digital human image frames are extracted through deep learning and graph convolutional networks. Inter-frame rendering collaboration rules for digital humans are generated, and frame-by-frame associated rendering operations are performed to generate an enhanced digital human image sequence with coherent inter-frame features.
The generated high-definition digital human videos are clearer and more realistic in terms of form and detail, and their movements are smoother and more natural, improving the quality of low-resolution digital human videos and bringing users a superior visual experience.
Smart Images

Figure CN121458812B_ABST
Abstract
Citation Information
Patent Citations
Dynamic video tactile feature extraction and rendering method
CN111291677A
Digital human video generation method based on multi-modal large model
CN120472059A