wondonghyeon/protest-detection-violence-estimation

Implementation of the model used in the paper Protest Activity Detection and Perceived Violence Estimation from Social Media Images (ACM Multimedia 2017)

48
/ 100
Emerging

Performs multi-task classification on protest imagery using a fine-tuned ImageNet ResNet50 backbone, simultaneously detecting protest presence, identifying visual attributes (signs, police, fire, group size), and predicting continuous violence perception scores. Built on PyTorch with training and batch inference pipelines supporting GPU acceleration, evaluated against the authors' curated UCLA Protest Image Dataset of 40K+ annotated images with 11 binary labels and violence regression targets.

185 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

185

Forks

46

Language

Jupyter Notebook

License

MIT

Last pushed

Mar 21, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/wondonghyeon/protest-detection-violence-estimation"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.