Spaces:
Running on Zero
Running on Zero
|
Download benchmark/README.md from Thiru1122/Multimodel_setup: direct link, hf CLI and curl.
- Browser
- Download file 1.64 kB
-
https://huggingface.co/spaces/Thiru1122/Multimodel_setup/resolve/main/benchmark/README.md
- Command line
-
hf download hf://spaces/Thiru1122/Multimodel_setup/benchmark/README.md
-
curl -L -o README.md https://huggingface.co/spaces/Thiru1122/Multimodel_setup/resolve/main/benchmark/README.md
1.64 kB
A newer version of the Gradio SDK is available: 6.32.0
Benchmark Dataset Directory
This directory provides an isolated, standardized benchmark harness for testing deepfake detectors.
Directory Structure
benchmark/
βββ README.md
βββ metadata.json
βββ images/
β βββ real/ # Ground-truth photographic images
β βββ ai_generated/ # Ground-truth synthetically generated images
β βββ manipulated/ # Ground-truth edited/spliced images
β βββ corrupted/ # Corrupted or malformed files for rejection testing
βββ results/ # Benchmark execution outputs (JSON and CSV)
Input Availability & Addition Protocol
In accordance with strict repository isolation rules:
- No files were copied from local workspaces or external paths outside this Space repository.
- No ground-truth labels were fabricated.
How to Manually Add Benchmark Images
- Place natural photographs into
benchmark/images/real/. - Place AI-generated images (e.g., from Diffusion, GAN, or Flow-Matching models) into
benchmark/images/ai_generated/. - Place edited or composite images into
benchmark/images/manipulated/. - Place truncated or corrupted test files into
benchmark/images/corrupted/. - Update
benchmark/metadata.jsonwith the filename, category, and verified ground-truth origin. If ground truth cannot be verified with documentary proof, markground_truth: "UNKNOWN".
Non-Authenticity Disclaimer
This benchmark framework is an operational diagnostic tool. Successful execution does not imply detector accuracy, and consensus across models does not verify photographic authenticity.