Zobrazeno 1 - 10
of 327
pro vyhledávání: '"Lap-Pui Chau"'
Publikováno v:
Journal of Remote Sensing, Vol 4 (2024)
Deep neural networks (DNNs) have found widespread applications in interpreting remote sensing (RS) imagery. However, it has been demonstrated in previous works that DNNs are susceptible and vulnerable to different types of noises, particularly advers
Externí odkaz:
https://doaj.org/article/83a12e93f65c4fe4b54868122a61cc69
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 36:2604-2612
To enhance low-light images to normally-exposed ones is highly ill-posed, namely that the mapping relationship between them is one-to-many. Previous works based on the pixel-wise reconstruction losses and deterministic processes fail to capture the c
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ef18016b928bcbcac98c34314a35ea4c
https://doi.org/10.1609/aaai.v36i3.20162
https://doi.org/10.1609/aaai.v36i3.20162
Publikováno v:
IEEE Transactions on Multimedia. 24:3835-3846
Stereo matching is a practical method to estimate depth information and retrieve 3D world in robot perception and autonomous driving scenarios. With the development of convolution neural networks (CNNs), deep-learning based stereo matching algorithms
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a945c2c35f36f78838016c83cbf4d64b
https://doi.org/10.1109/tmm.2021.3108900
https://doi.org/10.1109/tmm.2021.3108900
Publikováno v:
ACM Multimedia
Though convolutional neural networks are widely used in different tasks, lack of generalization capability in the absence of sufficient and representative data is one of the challenges that hinder their practical application. In this paper, we propos
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fc3488dc12afa6bf813c2e7572aba1dc
https://doi.org/10.1145/3474085.3475434
https://doi.org/10.1145/3474085.3475434
High Dynamic Range (HDR) imaging via multi-exposure fusion is an important task for most modern imaging platforms. In spite of recent developments in both hardware and algorithm innovations, challenges remain over content association ambiguities caus
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::aaaeb3920e6cefad9f1b75db7aa03a87
http://arxiv.org/abs/2107.06211
http://arxiv.org/abs/2107.06211
Publikováno v:
2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW).
The CenterTrack tracking algorithm achieves state-of-the-art tracking performance using a simple detection model and single-frame spatial offsets to localize objects and predict their associations in a single network. However, this joint detection an
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5c4f2b73c62db4bbf592c310dfad70cd
https://doi.org/10.1109/icmew53276.2021.9455993
https://doi.org/10.1109/icmew53276.2021.9455993
Publikováno v:
IEEE transactions on pattern analysis and machine intelligence. 44(10)
Coded aperture is a promising approach for capturing the 4-D light field (LF), in which the 4-D data are compressively modulated into 2-D coded measurements that are further decoded by reconstruction algorithms. The bottleneck lies in the reconstruct
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3fc3f215ad60e2321d940f5c19c55f02
https://pubmed.ncbi.nlm.nih.gov/34101585
https://pubmed.ncbi.nlm.nih.gov/34101585
Publikováno v:
IEEE transactions on image processing : a publication of the IEEE Signal Processing Society. 30
In this paper, we propose a novel self-training approach named Crowd-SDNet that enables a typical object detector trained only with point-level annotations (i.e., objects are labeled with points) to estimate both the center points and sizes of crowde
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::dc04e66f888ec2cdab518cb3ec18aa69
https://pubmed.ncbi.nlm.nih.gov/33539297
https://pubmed.ncbi.nlm.nih.gov/33539297
Vehicle re-identification (Re-ID) aims to retrieve images with the same vehicle ID across different cameras. Current part-level feature learning methods typically detect vehicle parts via uniform division, outside tools, or attention modeling. Howeve
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ef53fafe8e7ad1a685a18f03d6f4c82b
In this paper, we propose a simple yet effective crowd counting and localization network named SCALNet. Unlike most existing works that separate the counting and localization tasks, we consider those tasks as a pixel-wise dense prediction problem and
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2e7bba41fc36b7778624631dfd36f57b