Zobrazeno 1 - 10
of 131
pro vyhledávání: '"Peize Sun"'
Referring video object segmentation (R-VOS) is an emerging cross-modal task that aims to segment the target object referred by a language expression in all video frames. In this work, we propose a simple and unified framework built upon Transformer,
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::472842c1b24c77866bb2583ce66173ee
http://arxiv.org/abs/2201.00487
http://arxiv.org/abs/2201.00487
Autor:
Yifu Zhang, Peize Sun, Yi Jiang, Dongdong Yu, Fucheng Weng, Zehuan Yuan, Ping Luo, Wenyu Liu, Xinggang Wang
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783031200465
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fb5d2e9172eec975686073c03863eb7f
https://doi.org/10.1007/978-3-031-20047-2_1
https://doi.org/10.1007/978-3-031-20047-2_1
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783031198021
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3b1f0d8cb1ed7b0ea6c5fa91deabdb08
https://doi.org/10.1007/978-3-031-19803-8_43
https://doi.org/10.1007/978-3-031-19803-8_43
A typical pipeline for multi-object tracking (MOT) is to use a detector for object localization, and following re-identification (re-ID) for object association. This pipeline is partially motivated by recent progress in both object detection and re-I
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a4acc7f3da59180587c1e1532ccc6d8c
http://arxiv.org/abs/2111.14690
http://arxiv.org/abs/2111.14690
Publikováno v:
2021 IEEE/CVF International Conference on Computer Vision (ICCV).
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::297b1f1d415d935ad8e75d29bea417f4
https://doi.org/10.1109/iccv48922.2021.00807
https://doi.org/10.1109/iccv48922.2021.00807
Publikováno v:
2021 IEEE/CVF International Conference on Computer Vision (ICCV).
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ca1e58210fb381f0034bdca7690e17d0
https://doi.org/10.1109/iccv48922.2021.00865
https://doi.org/10.1109/iccv48922.2021.00865
Publikováno v:
IJCAI
This work presents a new fine-grained transparent object segmentation dataset, termed Trans10K-v2, extending Trans10K-v1, the first large-scale transparent object segmentation dataset. Unlike Trans10K-v1 that only has two limited categories, our new
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d6b39f209f01a4f0637a971996f16399
https://doi.org/10.24963/ijcai.2021/165
https://doi.org/10.24963/ijcai.2021/165
Publikováno v:
CVPR
In this paper, we introduce an anchor-box free and single shot instance segmentation method, which is conceptually simple, fully convolutional and can be used by easily embedding it into most off-the-shelf detection methods. Our method, termed PolarM
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e76050553baa3344901242547aad11eb
https://doi.org/10.1109/cvpr42600.2020.01221
https://doi.org/10.1109/cvpr42600.2020.01221
Autor:
Rufeng Zhang, Chenfeng Xu, Peize Sun, Wei Zhan, Tao Kong, Lei Li, Ping Luo, Masayoshi Tomizuka, Changhu Wang, Yi Jiang, Zehuan Yuan
Publikováno v:
CVPR
We present Sparse R-CNN, a purely sparse method for object detection in images. Existing works on object detection heavily rely on dense object candidates, such as $k$ anchor boxes pre-defined on all grids of image feature map of size $H\times W$. In
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3f7efc7672f853ff02ba73c542dc6f57
Autor:
Luo, Wei1,2,3 (AUTHOR) luowei@radi.ac.cn, Zhang, Guoqing1 (AUTHOR), Yuan, Quanbo1,4 (AUTHOR), Zhao, Yongxiang1 (AUTHOR), Chen, Hongce1 (AUTHOR), Zhou, Jingjie4,5 (AUTHOR), Meng, Zhaopeng4 (AUTHOR), Wang, Fulong1 (AUTHOR), Li, Lin1 (AUTHOR), Liu, Jiandong1 (AUTHOR), Wang, Guanwu1 (AUTHOR), Wang, Penggang1 (AUTHOR), Yu, Zhongde1 (AUTHOR)
Publikováno v:
PLoS ONE. 5/14/2024, Vol. 19 Issue 5, p1-22. 22p.