README.md

metadata

title: AnyAct User Study
colorFrom: blue
colorTo: indigo
sdk: gradio
python_version: '3.10'
app_file: app.py
pinned: false

AnyAct User Study

This repository contains a self-contained Gradio questionnaire for pairwise human motion reenactment evaluation.

Included Data

The Space-ready copy already bundles all 30 study cases inside this repository:

• videos/reference
• videos/anyact
• videos/vlm_hy_motion
• videos/echomotion

The app reads those files through data/study_config.json, so it no longer depends on directories outside the repository.

Runtime Storage

By default, the app stores runtime files under:

• results/responses.csv
• results/responses.jsonl
• results/participants
• results/plots

For Hugging Face Spaces, you should mount persistent storage and set:

USER_STUDY_RESULTS_DIR=/data/user_study_results

If the app is running inside a Space and /data exists, it automatically falls back to /data/user_study_results even without that variable.

Local Run

python -m venv .venv
.\.venv\Scripts\python.exe -m pip install -r requirements.txt
.\.venv\Scripts\python.exe app.py --share

Hugging Face Spaces Deployment

Option 1: Browser Upload

1. Create a new Hugging Face Space.
2. Choose Gradio as the SDK.
3. Upload the contents of this repository.
4. In the Space settings, optionally add persistent storage and set USER_STUDY_RESULTS_DIR=/data/user_study_results.

Option 2: CLI

hf auth login
deploy_to_hf_space.bat username/space-name

The helper script creates the Space if needed and uploads this repository while excluding .venv, .gradio, __pycache__, and results.

Files That Should Not Be Uploaded As Runtime Artifacts

The repository includes .gitignore rules so that local environments and transient results are not pushed by mistake:

• .venv
• .gradio
• __pycache__
• results

Analysis

To analyze collected responses:

.\.venv\Scripts\python.exe analyze_results.py

This writes:

• summary.csv
• participant_overview.csv
• plots/preference_barplot.png

inside the chosen results directory.