# Protest Activity Detection and Perceived Violence Estimation from Social Media Images

arXiv: https://arxiv.org/abs/1709.06204 · HTML: https://mlsanity.com/paper/1709.06204/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Sep 2017.

## Built on by (most cited first, 20 of 20)

- [Safety of Multimodal Large Language Models on Images and Texts](https://mlsanity.com/paper/2402.00357/) (2024)
- [UnsafeBench: Benchmarking Image Safety Classifiers on Real-World and AI-Generated Images](https://mlsanity.com/paper/2405.03486/) (2024)
- [ToViLaG: Your Visual-Language Generative Model is Also An Evildoer](https://mlsanity.com/paper/2312.11523/) (2023)
- [Deep Learning based Multi-Label Image Classification of Protest Activities](https://mlsanity.com/paper/2301.04212/) (2023)
- [The Reasonable Crowd: Towards evidence-based and interpretable models of driving behavior](https://mlsanity.com/paper/2107.13507/) (2021)
- [Ambiguous Images With Human Judgments for Robust Visual Event Classification](https://mlsanity.com/paper/2210.03102/) (2022)
- [Explaining Deep Convolutional Neural Networks via Latent Visual-Semantic Filter Attention](https://mlsanity.com/paper/2204.04601/) (2022)
- [Who Blames or Endorses Whom? Entity-to-Entity Directed Sentiment Extraction in News Text](https://mlsanity.com/paper/2106.01033/) (2021)
- [SafeVision: Efficient Image Guardrail with Robust Policy Adherence and Explainability](https://mlsanity.com/paper/2510.23960/) (2025)
- [Do You See What I See? Capabilities and Limits of Automated Multimedia Content Analysis](https://mlsanity.com/paper/2201.11105/) (2021)
- [Few-shot Learning for Multi-modal Social Media Event Filtering](https://mlsanity.com/paper/2211.10340/) (2022)
- [Improving Computer Vision Interpretability: Transparent Two-level Classification for Complex Scenes](https://mlsanity.com/paper/2407.03786/) (2024)
- [Leveraging Generative Artificial Intelligence for Causal Inference with Unstructured Data](https://mlsanity.com/paper/2507.03897/) (2025)
- [Zero-Shot Ranking Socio-Political Texts with Transformer Language Models to Reduce Close Reading Time](https://mlsanity.com/paper/2210.09179/) (2022)
- [VModA: An Effective Framework for Adaptive NSFW Image Moderation](https://mlsanity.com/paper/2505.23386/) (2025)
- [Short Video Uprising: How #BlackLivesMatter Content on TikTok Challenges the Protest Paradigm](https://mlsanity.com/paper/2206.09946/) (2022)
- [LLVMs4Protest: Harnessing the Power of Large Language and Vision Models for Deciphering Protests in the News](https://mlsanity.com/paper/2311.18241/) (2023)
- [Political Events using RAG with LLMs](https://mlsanity.com/paper/2502.15701/) (2025)
- [Protecting and Preserving Protest Dynamics for Responsible Analysis](https://mlsanity.com/paper/2604.05256/) (2026)
- [When Seeing Overrides Knowing: Visual Dominance and Deferral-Based Method for Personalized Safety in VLMs](https://mlsanity.com/paper/2609.04281/) (2026)

## Abstract (arXiv)

We develop a novel visual model which can recognize protesters, describe their activities by visual attributes and estimate the level of perceived violence in an image. Studies of social media and protests use natural language processing to track how individuals use hashtags and links, often with a focus on those items' diffusion. These approaches, however, may not be effective in fully characterizing actual real-world protests (e.g., violent or peaceful) or estimating the demographics of participants (e.g., age, gender, and race) and their emotions. Our system characterizes protests along these dimensions. We have collected geotagged tweets and their images from 2013-2017 and analyzed multiple major protest events in that period. A multi-task convolutional neural network is employed in order to automatically classify the presence of protesters in an image and predict its visual attributes, perceived violence and exhibited emotions. We also release the UCLA Protest Image Dataset, our novel dataset of 40,764 images (11,659 protest images and hard negatives) with various annotations of visual attributes and sentiments. Using this dataset, we train our model and demonstrate its effectiveness. We also present experimental results from various analysis on geotagged image data in several prevalent protest events. Our dataset will be made accessible at https://www.sscnet.ucla.edu/comm/jjoo/mm-protest/.