Zobrazeno 1 - 10
of 126
pro vyhledávání: '"Chiew-Lan Tai"'
Publikováno v:
IEEE Transactions on Visualization and Computer Graphics. 27:3745-3754
Sketches in existing large-scale datasets like the recent QuickDraw collection are often stored in a vector format, with strokes consisting of sequentially sampled points. However, most existing sketch recognition methods rasterize vector sketches as
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::df8354280e910302a09aa1d327ad8e46
https://doi.org/10.1109/tvcg.2020.2987626
https://doi.org/10.1109/tvcg.2020.2987626
Autor:
Zeyu Hu, Xuyang Bai, Jiaxiang Shang, Runze Zhang, Jiayu Dong, Xin Wang, Guangyuan Sun, Hongbo Fu, Chiew-Lan Tai
Publikováno v:
IEEE transactions on pattern analysis and machine intelligence.
In recent years, sparse voxel-based methods have become the state-of-the-arts for 3D semantic segmentation of indoor scenes, thanks to the powerful 3D CNNs. Nevertheless, being oblivious to the underlying geometry, voxel-based methods suffer from amb
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8e465f2a2504a5374842c0d7f082a6b1
https://pubmed.ncbi.nlm.nih.gov/35901000
https://pubmed.ncbi.nlm.nih.gov/35901000
Publikováno v:
IEEE Transactions on Circuits and Systems for Video Technology. 31:1738-1750
In this paper, we study the problem of multi-view sketch correspondence, where we take as input multiple freehand sketches with different views of the same object and predict as output the semantic correspondence among the sketches. This problem is c
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::29778e411725d0cc9addbcf358ef4c4d
https://doi.org/10.1109/tcsvt.2020.3015279
https://doi.org/10.1109/tcsvt.2020.3015279
Publikováno v:
IEEE Transactions on Visualization and Computer Graphics. 27:2355-2368
Specifying precise relationships among graphic elements is often a time-consuming process with traditional alignment tools. Automatic beautification of roughly designed layouts can provide a more efficient solution but often lead to undesired results
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d4f6f9d3d3136c55cd387122b4d75af7
https://doi.org/10.1109/tvcg.2019.2954321
https://doi.org/10.1109/tvcg.2019.2954321
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783031198113
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::6a8482eb61d3af0adfa5423253bf9ebf
https://doi.org/10.1007/978-3-031-19812-0_15
https://doi.org/10.1007/978-3-031-19812-0_15
Autor:
Jingyuan Liu, Nazmus Saquib, Zhutian Chen, Rubaiat Habib Kazi, Li-Yi Wei, Hongbo Fu, Chiew-Lan Tai
Videos are an accessible form of media for analyzing sports postures and providing feedback to athletes. Existing sport-specific systems embed bespoke human pose attributes and thus can be hard to scale for new attributes, especially for users withou
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::135f92eb0a868643c806591b1d1ded46
Publikováno v:
2021 IEEE/CVF International Conference on Computer Vision (ICCV).
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d21ea156c26efb5e769b5d669eca9e23
https://doi.org/10.1109/iccv48922.2021.01132
https://doi.org/10.1109/iccv48922.2021.01132
Publikováno v:
IEEE Computer Graphics and Applications. 39:38-51
We present a simple and efficient method based on deep learning to automatically decompose sketched objects into semantically valid parts. We train a deep neural network to transfer existing segmentations and labelings from three-dimensional (3-D) mo
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::80be535a86ee7bc86473c556ad4d59e9
https://doi.org/10.1109/mcg.2018.2884192
https://doi.org/10.1109/mcg.2018.2884192
Publikováno v:
CVPR
Removing outlier correspondences is one of the critical steps for successful feature-based point cloud registration. Despite the increasing popularity of introducing deep learning methods in this field, spatial consistency, which is essentially estab
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::370ab0c5e78cef317a520acd368f9164
http://arxiv.org/abs/2103.05465
http://arxiv.org/abs/2103.05465
Autor:
Zeyu Hu, Xuyang Bai, Jiaxiang Shang, Runze Zhang, Jiayu Dong, Xin Wang, Guangyuan Sun, Hongbo Fu, Chiew-Lan Tai
In recent years, sparse voxel-based methods have become the state-of-the-arts for 3D semantic segmentation of indoor scenes, thanks to the powerful 3D CNNs. Nevertheless, being oblivious to the underlying geometry, voxel-based methods suffer from amb
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ef9083927569a865727c59b6e93d6476