Learning to generate video object segment proposals

Abstract

This paper proposes a fully automatic pipeline to generate accurate object segment proposals in realistic videos. Our approach first detects generic object proposals for all video frames and then learns to rank them using a Convolutional Neural Networks (CNN) descriptor built on appearance and motion cues. The ambiguity of the proposal set can be reduced… (More)
DOI: 10.1109/ICME.2017.8019535

Topics

7 Figures and Tables

Slides referencing similar topics