OVOS Occluded Video Object Segmentation dataset

Citation Author(s):
Bo
Miao
University of Western Australia
Mohammed
Bennamoun
University of Western Australia
Yongsheng
Gao
Griffith University
Ajmal
Mian
University of Western Australia
Submitted by:
bo miao
Last updated:
Mon, 07/18/2022 - 03:14
DOI:
10.21227/6w7a-qq88
Data Format:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

Semi-supervised video object segmentation aims to leverage the ground truth object masks given for the first frame to segment video sequences at the pixel level. OVOS is a dataset to evaluate the performance of video object segmentation under occlusions.

Instructions: 

--JPEGImages (Video frames)

--Annotations (Ground truth masks of video frames)

--ImageSets (Video names)

Funding Agency: 
Australian Research Council