Detecting events and key actors in multi-person videos

TL;DR

This paper introduces a novel attention-based model for multi-person video event detection that automatically identifies key actors without explicit annotations, demonstrating superior performance on a new basketball dataset.

Contribution

The paper presents an attention mechanism integrated with RNNs for detecting events and key actors in multi-person videos without requiring explicit annotations.

Findings

Outperforms state-of-the-art methods in event classification and detection

Successfully localizes relevant players using attention mechanism

Introduces a new basketball dataset with 14K event annotations

Abstract

Multi-person event recognition is a challenging task, often with many people active in the scene but only a small subset contributing to an actual event. In this paper, we propose a model which learns to detect events in such videos while automatically "attending" to the people responsible for the event. Our model does not use explicit annotations regarding who or where those people are during training and testing. In particular, we track people in videos and use a recurrent neural network (RNN) to represent the track features. We learn time-varying attention weights to combine these features at each time-instant. The attended features are then processed using another RNN for event detection/classification. Since most video datasets with multiple people are restricted to a small number of videos, we also collected a new basketball dataset comprising 257 basketball games with 14K event annotations corresponding to 11 event classes. Our model outperforms state-of-the-art methods for both event classification and detection on this new dataset. Additionally, we show that the attention mechanism is able to consistently localize the relevant players.