Making a long story short: A Multi-Importance fast-forwarding egocentric videos with the emphasis on relevant objects

TL;DR

This paper introduces MIFF, a fully automatic method for fast-forwarding egocentric videos that emphasizes relevant objects based on user preferences, significantly increasing semantic content preservation.

Contribution

The paper presents a novel Multi-Importance Fast-Forward approach that balances smoothness and relevance, with a learning process tailored to user preferences for semantic emphasis.

Findings

MIFF retains over 3 times more semantic content than previous methods.

The approach adapts to user preferences for defining semantic importance.

A discussion on stabilization techniques for fast-forward egocentric videos is included.

Abstract

The emergence of low-cost high-quality personal wearable cameras combined with the increasing storage capacity of video-sharing websites have evoked a growing interest in first-person videos, since most videos are composed of long-running unedited streams which are usually tedious and unpleasant to watch. State-of-the-art semantic fast-forward methods currently face the challenge of providing an adequate balance between smoothness in visual flow and the emphasis on the relevant parts. In this work, we present the Multi-Importance Fast-Forward (MIFF), a fully automatic methodology to fast-forward egocentric videos facing these challenges. The dilemma of defining what is the semantic information of a video is addressed by a learning process based on the preferences of the user. Results show that the proposed method keeps over $3$ times more semantic content than the state-of-the-art fast-forward. Finally, we discuss the need of a particular video stabilization technique for fast-forward egocentric videos.