TY - GEN
T1 - Unsupervised Video Object Segmentation Using Motion Saliency-Guided Spatio-Temporal Propagation
AU - Hu, Yuan Ting
AU - Huang, Jia Bin
AU - Schwing, Alexander G.
N1 - Funding Information:
Acknowledgments. This material is based upon work supported in part by the National Science Foundation under Grant No. 1718221, 1755785, Samsung, and 3M. We thank NVIDIA for providing the GPUs used for this research.
Publisher Copyright:
© 2018, Springer Nature Switzerland AG.
PY - 2018
Y1 - 2018
N2 - Unsupervised video segmentation plays an important role in a wide variety of applications from object identification to compression. However, to date, fast motion, motion blur and occlusions pose significant challenges. To address these challenges for unsupervised video segmentation, we develop a novel saliency estimation technique as well as a novel neighborhood graph, based on optical flow and edge cues. Our approach leads to significantly better initial foreground-background estimates and their robust as well as accurate diffusion across time. We evaluate our proposed algorithm on the challenging DAVIS, SegTrack v2 and FBMS-59 datasets. Despite the usage of only a standard edge detector trained on 200 images, our method achieves state-of-the-art results outperforming deep learning based methods in the unsupervised setting. We even demonstrate competitive results comparable to deep learning based methods in the semi-supervised setting on the DAVIS dataset.
AB - Unsupervised video segmentation plays an important role in a wide variety of applications from object identification to compression. However, to date, fast motion, motion blur and occlusions pose significant challenges. To address these challenges for unsupervised video segmentation, we develop a novel saliency estimation technique as well as a novel neighborhood graph, based on optical flow and edge cues. Our approach leads to significantly better initial foreground-background estimates and their robust as well as accurate diffusion across time. We evaluate our proposed algorithm on the challenging DAVIS, SegTrack v2 and FBMS-59 datasets. Despite the usage of only a standard edge detector trained on 200 images, our method achieves state-of-the-art results outperforming deep learning based methods in the unsupervised setting. We even demonstrate competitive results comparable to deep learning based methods in the semi-supervised setting on the DAVIS dataset.
UR - http://www.scopus.com/inward/record.url?scp=85055122590&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85055122590&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-01246-5_48
DO - 10.1007/978-3-030-01246-5_48
M3 - Conference contribution
AN - SCOPUS:85055122590
SN - 9783030012458
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 813
EP - 830
BT - Computer Vision – ECCV 2018 - 15th European Conference, 2018, Proceedings
A2 - Hebert, Martial
A2 - Ferrari, Vittorio
A2 - Sminchisescu, Cristian
A2 - Weiss, Yair
PB - Springer
T2 - 15th European Conference on Computer Vision, ECCV 2018
Y2 - 8 September 2018 through 14 September 2018
ER -