This is the template for the Cryptography and Information Security Conference (CISC).
Due to copyright concerns about the BiauKai font, this template uses AR PL UKai TW as the main CJK font instead. But you're still able to upload the font to Overleaf by yourself.
Notice: \nocite{*} is used to display reference examples. You should delete this line.
The default template is for Chinese paper, please change the following parameters to the English version.
labelsep = period
% English = period, Chinese = space
\renewcommand{\Authsep}{~}
\renewcommand{\Authand}{~}
\renewcommand{\Authands}{~}
\renewcommand\figurename{圖}
\renewcommand\tablename{表}
\renewcommand\refname{參考文獻}
% comment out these lines for English paper
\parindent=1em
% English = 1em, Chinese = 2em
12th Edition of the Language Resources and Evaluation Conference LaTeX template.
Source: https://lrec2020.lrec-conf.org/en/submission2020/authors-kit/.
Paper presented at ICCV 2019.
This paper targets the task with discrete and periodic
class labels (e.g., pose/orientation estimation) in the context of deep learning. The commonly used cross-entropy or
regression loss is not well matched to this problem as they
ignore the periodic nature of the labels and the class similarity, or assume labels are continuous value. We propose to
incorporate inter-class correlations in a Wasserstein training framework by pre-defining (i.e., using arc length of a
circle) or adaptively learning the ground metric. We extend
the ground metric as a linear, convex or concave increasing
function w.r.t. arc length from an optimization perspective.
We also propose to construct the conservative target labels
which model the inlier and outlier noises using a wrapped
unimodal-uniform mixture distribution. Unlike the one-hot
setting, the conservative label makes the computation of
Wasserstein distance more challenging. We systematically
conclude the practical closed-form solution of Wasserstein
distance for pose data with either one-hot or conservative
target label. We evaluate our method on head, body, vehicle and 3D object pose benchmarks with exhaustive ablation studies. The Wasserstein loss obtaining superior performance over the current methods, especially using convex mapping function for ground metric, conservative label,
and closed-form solution.
Xiaofeng Liu, Yang Zou, Tong Che, Peng Ding, Ping Jia, Jane You, B.V.K. Vijaya Kumar
This is an UPDATED template suitable for submissions to the 14th Conference on Spatial Information Theory, which will be hosted by the Chair for Information Science at the University of Regensburg, Germany. It is a provided as a means of making things easier for those who might not be too familiar with writing LaTeX. The project uses NOW THE 2019 VERSION of the LIPICs class file. Submissions to the conference must adhere to the official LIPICs guidelines for authors to be found at: http://drops.dagstuhl.de/styles/lipics-v2019/lipics-v2019-authors.zip.