Zobrazeno 1 - 8
of 8
pro vyhledávání: '"Dezhao Luo"'
Publikováno v:
ACM Transactions on Multimedia Computing, Communications, and Applications. 18:1-21
Existing video self-supervised learning methods mainly rely on trimmed videos for model training. They apply their methods and verify the effectiveness on trimmed video datasets including UCF101 and Kinetics-400, among others. However, trimmed datase
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2737f1125e1de055394aa46ab0ec6b65
https://doi.org/10.1145/3473342
https://doi.org/10.1145/3473342
Publikováno v:
IEEE Transactions on Neural Networks and Learning Systems. :1-15
In this study, we propose a novel pretext task and a self-supervised motion perception (SMP) method for spatiotemporal representation learning. The pretext task is defined as video playback rate perception, which utilizes temporal dilated sampling to
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::33f700bd32a334a45e16ed52edd6e900
https://doi.org/10.1109/tnnls.2022.3160860
https://doi.org/10.1109/tnnls.2022.3160860
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783031159367
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::cfa5dce4039484a3d59d6e5e58c4b1ad
https://doi.org/10.1007/978-3-031-15937-4_43
https://doi.org/10.1007/978-3-031-15937-4_43
Publikováno v:
CVPR
In self-supervised spatio-temporal representation learning, the temporal resolution and long-short term characteristics are not yet fully explored, which limits representation capabilities of learned models. In this paper, we propose a novel self-sup
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::efcd02b8996af844d45c50b12fd8b732
https://doi.org/10.1109/cvpr42600.2020.00658
https://doi.org/10.1109/cvpr42600.2020.00658
Publikováno v:
AAAI
We propose a novel self-supervised method, referred to as Video Cloze Procedure (VCP), to learn rich spatial-temporal representations. VCP first generates "blanks" by withholding video clips and then creates "options" by applying spatio-temporal oper
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c6e4b4fb8814a89d014894047612d678
http://arxiv.org/abs/2001.00294
http://arxiv.org/abs/2001.00294
Publikováno v:
CVPR Workshops
Geometric shape understanding provides an intuitive representation of object shapes. Skeleton is typical geometrical information. Lots of traditional approaches are developed for skeleton extraction and pruning, while it is still a new area to invest
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::5ff9bb4a4d11e6f9eb4867f712023b06
https://doi.org/10.1109/cvprw.2019.00154
https://doi.org/10.1109/cvprw.2019.00154
Publikováno v:
CVPR Workshops
We propose an efficient and effective control point extraction algorithm for parametric skeleton generation. The object skeleton pixels are predicted via an hourglass network and partitioned into skeleton branches using Gaussian Mixture Models. For e
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e8d1a1d9aa759b588490adf0ce5af920
https://doi.org/10.1109/cvprw.2019.00153
https://doi.org/10.1109/cvprw.2019.00153
Autor:
Dezhao Luo
Publikováno v:
IOP Conference Series: Materials Science and Engineering. 768:042044
Based on the enclosed vibrating screen and elastic cantilever screen as the research object, through the enclosed vibrating screen and elastic cantilever vibrating screen structure characteristics and screening process of research, for railway mainte
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fc942a7bea5893a7e106eef65b54ad55
https://doi.org/10.1088/1757-899x/768/4/042044
https://doi.org/10.1088/1757-899x/768/4/042044