OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Deep Tempest

Deep Tempest is a research project that uses deep-learning restoration to improve images reconstructed from unintended electromagnetic emissions from HDMI displays, extending the gr-tempest approach.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Deep Tempest — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when studying deep-learning restoration of images recovered from HDMI electromagnetic emissions, or when working with its described research data and capture-to-inference workflow.

Who it is for

  • Researchers investigating TEMPEST-style image recovery and restoration.
  • Practitioners comparing deep-learning restoration with unmodified gr-tempest output.
  • Researchers seeking synthetic and real capture data for experiments, training or evaluation.

Documented uses

  • Improve the visual quality of images recovered from HDMI electromagnetic captures.
  • Run inference and train a DRUNet-based restoration architecture using the repository's workflow guides.
  • Compare restoration results with the original gr-tempest output.
  • Use the published synthetic and captured images in research experiments.
Before you choose

Requirements & limits

The source repository is publicly accessible. No license is supplied in the repository evidence.

  • The project covers a specialized signal-recovery and image-restoration workflow, not general-purpose image enhancement.
  • The repository separates gr-tempest capture work from end-to-end inference and training guides.
  • The dataset download is approximately 7 GB and includes synthetic and real captured images.
  • Setup requires Tesseract OCR and specifies PyTorch/Torchvision CUDA 12.1 packages.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare the Deep Tempest Python environment

Set up the documented Python dependencies before proceeding to the repository's separate capture, inference or training instructions.

Before you start

  • Python 3.10 or 3.12; the setup examples use Python 3.12.
  • Tesseract OCR installed before environment setup.
  • Choose Conda or Pyenv with venv.

Steps

  1. Install Tesseract OCR using the system package manager.

  2. Create and activate a Conda environment with Python 3.12, or create and activate a Python 3.12 venv.

  3. For Conda, update the environment from tempest_conda.yml; for venv, install dependencies from tempest_pyenv.txt.

  4. Install the specified PyTorch and Torchvision CUDA 12.1 builds, pybind11 2.13.6 and fastwer from the documented Git repository.

Expected outcome

A Python environment prepared according to one of the repository's two setup paths.

Guide limitations

  • Environment setup does not itself capture signals, run inference or train the model; those stages have separate guides in the repository.
  • The README identifies Ubuntu 22.04.5 LTS and 24.04.2 LTS as tested systems and states that the environment uses CUDA 12.1 PyTorch builds.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
7 November 2026
Original website title
emidan19/deep-tempest · Official homepage title · retained in its source language
How we review sourcesSuggest a correction