Zobrazeno 1 - 10
of 13
pro vyhledávání: '"Daiki Ikami"'
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 36:3161-3169
Unsupervised domain adaptation (UDA) has been highly successful in transferring knowledge acquired from a label-rich source domain to a label-scarce target domain. Open-set domain adaptation (open-set DA) and universal domain adaptation (UniDA) have
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::9e8529e05c36a0fbfff99a486335e79d
https://doi.org/10.1609/aaai.v36i3.20224
https://doi.org/10.1609/aaai.v36i3.20224
Rotation is frequently listed as a candidate for data augmentation in contrastive learning but seldom provides satisfactory improvements. We argue that this is because the rotated image is always treated as either positive or negative. The semantics
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e861e156cd8e5152effc6229bbcff070
Publikováno v:
IJCAI
Lifelong learning aims to train a highly expressive model for a new task while retaining all knowledge for previous tasks. However, many practical scenarios do not always require the system to remember all of the past knowledge. Instead, ethical cons
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::24d182258602d3fe2a03d68d61eb99bc
https://doi.org/10.24963/ijcai.2021/137
https://doi.org/10.24963/ijcai.2021/137
Publikováno v:
CVPR
Many variants of unsupervised domain adaptation (UDA) problems have been proposed and solved individually. Its side effect is that a method that works for one variant is often ineffective for or not even applicable to another, which has prevented pra
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6ee8349114c6fd6932fe225bf1f0d94d
http://arxiv.org/abs/2106.01656
http://arxiv.org/abs/2106.01656
Publikováno v:
ICPR
We propose a new optimization framework for aleatoric uncertainty estimation in regression problems. Existing methods can quantify the error in the target estimation, but they tend to underestimate it. To obtain the predictive uncertainty inherent in
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5216d682e799dca61d34a0d1ef581990
https://doi.org/10.1109/icpr48806.2021.9412324
https://doi.org/10.1109/icpr48806.2021.9412324
Publikováno v:
WACV
Weight Normalization (WN) is an essential building block in deep learning. However, even state-of-the-art WN methods need to be combined with activation normalization methods, such as Batch Normalization (BN), to provide the same classification accur
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::b722e9f2a6bc4b230295ebf0fb18543e
https://doi.org/10.1109/wacv48630.2021.00265
https://doi.org/10.1109/wacv48630.2021.00265
Publikováno v:
Computer Vision – ACCV 2020 ISBN: 9783030695347
ACCV (3)
ACCV (3)
We study the shape of the convolution kernels in the upsampling block for deep monocular depth estimation. First, our empirical analysis shows that the depth estimation accuracy can be improved consistently by only changing the shape of the two conse
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::1c5294ee3617d4ebaa9e1954637694cb
https://doi.org/10.1007/978-3-030-69535-4_27
https://doi.org/10.1007/978-3-030-69535-4_27
Publikováno v:
Computer Vision – ECCV 2020 ISBN: 9783030586096
ECCV (12)
ECCV (12)
Semi-supervised learning (SSL) has been proposed to leverage unlabeled data for training powerful models when only limited labeled data is available. While existing SSL methods assume that samples in the labeled and unlabeled data share the classes o
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::427b9519c1f158b01c0fdcee2eb5181a
https://doi.org/10.1007/978-3-030-58610-2_26
https://doi.org/10.1007/978-3-030-58610-2_26
Publikováno v:
ISM
Manga or Japanese comics are a popular medium and their images comprise line drawings and screentones. This study investigates the screentone synthesis task that involves translation from line drawings to manga images. Screentones have regular patter
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::b0a65d0006cbbd8c79d9690bdb9037ed
https://doi.org/10.1109/ism46123.2019.00046
https://doi.org/10.1109/ism46123.2019.00046
Publikováno v:
CVPR
The goal of graph-based clustering is to divide a dataset into disjoint subsets with members similar to each other from an affinity (similarity) matrix between data. The most popular method of solving graph-based clustering is spectral clustering. Ho
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::bb0e6a871e758300540143c6f8f904d5
https://doi.org/10.1109/cvpr.2018.00364
https://doi.org/10.1109/cvpr.2018.00364