Search Results for author: Arsalan Heydarian

Found 3 papers, 1 papers with code

WcDT: World-centric Diffusion Transformer for Traffic Scene Generation

1 code implementation2 Apr 2024 Chen Yang, Aaron Xuxiang Tian, Dong Chen, Tianyu Shi, Arsalan Heydarian

To enhance the scene diversity and stochasticity, the historical trajectory data is first preprocessed and encoded into latent space using Denoising Diffusion Probabilistic Models (DDPM) enhanced with Diffusion with Transformer (DiT) blocks.

Autonomous Driving Denoising +1

Building Performance Simulations Can Inform IoT Privacy Leaks in Buildings

no code implementations26 Mar 2023 Alan Wang, Bradford Campbell, Arsalan Heydarian

Specifically, in this proof-of-concept exploration, we demonstrate the potential of physics-based simulation models to quantify the minimal number of positions necessary to capture sensitive inferences.

Cannot find the paper you are looking for? You can Submit a new open access paper.