Meirui Wang

Deep learning-based Synthetic Aperture Radar (SAR) despeckling faces two key challenges: physical model mismatch, where simplified noise assumptions fail to capture speckle’s statistics, and architectural generality, where standard networks struggle to separate noise from texture. To address this, we propose DA-PhysDiff, a dual-aware physics-informed diffusion model for SAR image despeckling. The proposed LogGamma-based stochastic differential equation models multiplicative speckle more accurately, while the dual-aware network composed of a Statistics-Conditioned Gamma Attention (SCGA) and a Frequency-Refined Adaptive Wavelet Convolution (FRA-WTConv) enhances both distribution and frequency awareness. Experiments on HRSID, MiniSAR, and synthetic benchmarks demonstrate that DA-PhysDiff achieves up to +2.5 dB PSNR and +4.8% EPD-ROA improvement over recent diffusion models, while reducing parameters by 19% compared with SAR-DDPM and maintaining real-time inference.

To address traditional VR real-time rendering limitations—single-dimensional precision control, poor multimodal interaction-rendering coordination, and computational power-image quality imbalance in complex scenes—we propose a deep learning-based multimodal fusion algorithm. It forms an input-fusion-decision-rendering-feedback loop: capturing gesture/audio via VR headsets’ sensors, purifying features, fusing with a lightweight Transformer, using an interaction intensity-viewing distance attention system for high/medium/low precision, and combining local NeRF, LoD, GPU protection to ensure ≥90fps. Timestamp-aligned multimodal feedback is added. Experiments show 35%+ more target details than traditional LoD, average GPU utilization ≤78%, aiding VR education, gaming, and offices.