Tracking Large-Scale Video Remix in Real-World Events

TL;DR

This paper introduces a scalable method for detecting and analyzing visual memes in large-scale videos, revealing insights into latent interactions and content influence during major news events.

Contribution

It develops high-accuracy detection algorithms for visual memes, combines them with text analysis, and models their influence and popularity on YouTube at scale.

Findings

High accuracy in visual meme detection.

Joint model outperforms concurrence model.

Predicts meme volume with ~2% error and lifespan with ~17% error.

Abstract

Social information networks, such as YouTube, contains traces of both explicit online interaction (such as "like", leaving a comment, or subscribing to video feed), and latent interactions (such as quoting, or remixing parts of a video). We propose visual memes, or frequently re-posted short video segments, for tracking such latent video interactions at scale. Visual memes are extracted by scalable detection algorithms that we develop, with high accuracy. We further augment visual memes with text, via a statistical model of latent topics. We model content interactions on YouTube with visual memes, defining several measures of influence and building predictive models for meme popularity. Experiments are carried out on with over 2 million video shots from more than 40,000 videos on two prominent news events in 2009: the election in Iran and the swine flu epidemic. In these two events, a high percentage of videos contain remixed content, and it is apparent that traditional news media and citizen journalists have different roles in disseminating remixed content. We perform two quantitative evaluations for annotating visual memes and predicting their popularity. The joint statistical model of visual memes and words outperform a concurrence model, and the average error is ~2% for predicting meme volume and ~17% for their lifespan.