Learning Video Object Segmentation from Static Images

TL;DR

This paper presents a novel video object segmentation method that leverages static image training and combines offline and online learning to achieve accurate, flexible segmentation across various datasets and annotation types.

Contribution

It introduces a guided instance segmentation approach that operates per frame, using static image-trained convnets and combined learning strategies for improved video segmentation.

Findings

Achieves competitive results on three datasets

Handles multiple annotation types effectively

Operates with static image training only

Abstract

Inspired by recent advances of deep learning in instance segmentation and object tracking, we introduce video object segmentation problem as a concept of guided instance segmentation. Our model proceeds on a per-frame basis, guided by the output of the previous frame towards the object of interest in the next frame. We demonstrate that highly accurate object segmentation in videos can be enabled by using a convnet trained with static images only. The key ingredient of our approach is a combination of offline and online learning strategies, where the former serves to produce a refined mask from the previous frame estimate and the latter allows to capture the appearance of the specific object instance. Our method can handle different types of input annotations: bounding boxes and segments, as well as incorporate multiple annotated frames, making the system suitable for diverse applications. We obtain competitive results on three different datasets, independently from the type of input annotation.