High-resolution efficient image generation from WiFi CSI using a pretrained latent diffusion model

TL;DR

LatentCSI is a new method that generates high-resolution images of physical environments from WiFi CSI data using a pretrained latent diffusion model, offering efficiency, quality, and text-guided control.

Contribution

It introduces a lightweight neural network to map WiFi CSI to a latent space of a pretrained diffusion model, bypassing pixel-space generation and enabling efficient, high-quality image synthesis with text guidance.

Findings

Outperforms comparable baselines in efficiency and perceptual quality

Works effectively on real WiFi CSI datasets and public benchmarks

Provides practical text-guided controllability in image generation

Abstract

We present LatentCSI, a novel method for generating images of the physical environment from WiFi CSI measurements that leverages a pretrained latent diffusion model (LDM). Unlike prior approaches that rely on complex and computationally intensive techniques such as GANs, our method employs a lightweight neural network to map CSI amplitudes directly into the latent space of an LDM. We then apply the LDM's denoising diffusion model to the latent representation with text-based guidance before decoding using the LDM's pretrained decoder to obtain a high-resolution image. This design bypasses the challenges of pixel-space image generation and avoids the explicit image encoding stage typically required in conventional image-to-image pipelines, enabling efficient and high-quality image synthesis. We validate our approach on two datasets: a wide-band CSI dataset we collected with off-the-shelf WiFi devices and cameras; and a subset of the publicly available MM-Fi dataset. The results demonstrate that LatentCSI outperforms baselines of comparable complexity trained directly on ground-truth images in both computational efficiency and perceptual quality, while additionally providing practical advantages through its unique capacity for text-guided controllability.