Publications
Head-Related Transfer Function Interpolation with a Spherical CNN
We propose a spherical convolutional neural network method for HRTF interpolation. The proposed method realizes the convolution process by decomposing and reconstructing HRTF through the Spherical Harmonics (SHs). The SHs, an orthogonal function set defined on a sphere, allow the convolution layers to effectively capture the spatial features of HRTFs that are sampled on a sphere.
Physics Informed Neural Network for Head-Related Transfer Function Upsampling
We propose a physics-informed neural network (PINN) method for HRTF upsampling. The PINN exploits the Helmholtz equation, the governing equation of acoustic wave propagation, for regularizing the upsampling process. This helps the generation of physically valid upsamplings that generalize beyond the measured HRTF. Furthermore, the size (width and depth) of the PINN is set according to the Helmholtz equation and its solutions, the spherical harmonics (SHs). This makes the PINN have an appropriate level of expressive power and thus does not suffer from the over-fitting problem.
Sound Field Estimation around a Rigid Sphere with Physics-informed Neural Network
We propose a method for sound field estimation based on a physics-informed neural network. This method integrates physical knowledge into the architecture and training process of the network.
Closed-Form Error Propagation on SEn(3) Group for Invariant EKF With Applications to VINS
We establish the closed-form formula for the error propagation for the Invariant extended Kalman filter (IEKF) in the presence of random noises and apply it to vision-aided inertial navigation. Moreover, we use the theoretic results to add the compensation parts for IEKF.