Video captioning with recurrent networks based on frame- and video-level features and visual content classification

TL;DR

This paper presents a video captioning system using recurrent neural networks that combines frame-level features, video-specific features, and visual content classifiers to generate descriptive text for short video clips.

Contribution

It extends static image captioning models to videos by integrating multiple feature types and content classifiers, improving captioning performance.

Findings

Combining keyframe features, dense trajectory features, and content classifiers yields better results.

Utilizing multiple feature sources enhances caption quality over individual features.

The system participated successfully in the Large Scale Movie Description Challenge 2015.

Abstract

In this paper, we describe the system for generating textual descriptions of short video clips using recurrent neural networks (RNN), which we used while participating in the Large Scale Movie Description Challenge 2015 in ICCV 2015. Our work builds on static image captioning systems with RNN based language models and extends this framework to videos utilizing both static image features and video-specific features. In addition, we study the usefulness of visual content classifiers as a source of additional information for caption generation. With experimental results we show that utilizing keyframe based features, dense trajectory video features and content classifier outputs together gives better performance than any one of them individually.