Hugging Face's logo

Spaces:
Jomaric
/
visual_content_labeler
Runtime error

App Files Community
visual_content_labeler / README.md
Jomaric's picture
Jomaric
feat: initial release of vision social research space
b27edc9
preview code
|
Raw
History Blame Contribute Delete
1.05 kB

A newer version of the Gradio SDK is available: 6.20.0

Upgrade
metadata 
title: Visual Content Labeler
emoji: 🕸️
colorFrom: green
colorTo: indigo
sdk: gradio
app_file: app.py
pinned: false

Visual Content Labeler & Batch Database Builder

An interactive computational tool designed for social science researchers, media scholars, and digital humanities students to expedite labeling image archives and building quantitative database sheets.

Features

  1. Batch Uploading: Upload a ZIP archive containing hundreds of images or select multiple image files directly.
  2. Automated Face Counting: Instantly locates and counts human faces using localized, zero-overhead Haar-Cascade classifiers.
  3. Automated Object Suggestion: Suggests candidate labels using a lightweight pre-trained MobileNetV3 convolutional neural network.
  4. Interactive Review Gallery: Step through the uploaded images, view bounding boxes, and edit or append your own qualitative codes.
  5. One-Click Export: Keep track of progress in a live spreadsheet and export your finished work as a standardized, clean CSV dataset.