基于图像编码器的头相关传递函数个性化方法
By using an image encoder-based method to predict the amplitude and phase of HRTF using user ear images and physiological parameters, the problem of high cost and lack of phase in obtaining personalized HRTF is solved, realizing the generation and application of personalized spatial audio.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUILIN UNIV OF ELECTRONIC TECH
- Filing Date
- 2024-03-15
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, methods for obtaining personalized head-related transfer functions (HRTFs) are costly and lack phase information, making it impossible to directly generate personalized spatial audio.
Using an image encoder-based approach, the method utilizes user ear images and some human physiological parameters as input. A nonlinear mapping relationship is established through an autoencoder to predict the amplitude and phase of the HRTF, generate a personalized HRTF, and convert it into a time-domain HRIR, which can be directly used in virtual acoustic products.
It effectively reduces the cost of measuring human physiological parameters, and the generated HRTF can be directly used in virtual acoustic products to achieve a personalized spatial audio immersion and meet the needs of a wide range of users.
Smart Images
Figure CN118135235B_ABST