D. Han

PedestrianDiffusion: Multimodal Generative Denoising and Dense State Estimation for Inertial Navigation

This work proposes PedestrianDiffusion, a multimodal spectral-domain generative framework reformulating dense 6D state estimation as a continuous conditional denoising process, and introduces a zero-shot semantic conditioning mechanism leveraging vision-language embeddings as categorical priors to generalize across heterogeneous sensor noise profiles.