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.
Sample document using the v2019 style version of the series OASIcs - OpenAccess Series in Informatics. OASIcs offers a venue for the Open Access and online publication of peer-reviewed proceedings based on international scientific events (workshops, symposia, conferences, ...) that took place outside of Schloss Dagstuhl.
See https://www.dagstuhl.de/oasics
https://github.com/dagstuhl-publishing/styles/
This is a sample file for legacy SIGCHI extended abstracts, using acmart.cls v1.71 (2020/05/04).
Starting in Spring 2020 ACM retired SIGCHI Extended Abstract format (sigchi-a). ACM will not, under any circumstances, accept documents in this format for publication and will not offer technical support to the authors who use this template.
You may use this format in the nonacm mode only, as in \documentclass[sigchi-a, nonacm]{acmart}.
These instructions provide guidelines for preparing papers for AIAA Technical Journals using LaTeX. You may also use it for preparing papers for AIAA conferences by toggling the documentclass option in the template.
AIAA journals provide a panoramic journey from yesterday’s challenges through today's most important aerospace advances in research and development.AIAA’s original research papers present concepts, methods of analysis, technical knowledge, exploratory developments, and new applications.
To begin writing online (in your browser), simply click the Open as Template button, above. Additional guidelines for preparing your submission are included within the template itself.
If you'd like to download any of the template files including the .cls file, please click "Open as template" above, then download the template “Source” zip file from the menu.
This template is designed for submissions to all current AIAA journals:
AIAA Journal
Journal of Aerospace Information Systems
Journal of Aircraft
Journal of Air Transportation
Journal of Guidance, Control and Dynamics
Journal of Propulsion and Power
Journal of Spacecraft and Rockets
Journal of Thermophysics and Heat Transfer
If you're new to Overleaf and LaTeX, check out our free introductory course for help getting started.
These instructions provide guidelines for preparing papers for AIAA Technical Conferences using LaTeX.
AIAA is the catalyst for inspired idea exchange and solutions, a convener of the most original perspectives, and curator of essential research information. For the past 50 years, individuals and teams from around the globe have presented their latest research to their peers at AIAA conferences.
To begin writing online (in your browser), simply click the Open as Template button, above. Additional guidelines for preparing your submission are included within the template itself.
If you'd like to download any of the template files including the .cls file, please click "Open as template" above, then download the template “Source” zip file from the menu.
For a list of AIAA forums and other events currently accepting abstracts, visit the AIAA events listing page.
If you're new to Overleaf and LaTeX, check out our free introductory course for help getting started.
Exemple de MCOT pour les TIPE (Travaux d'Initiative Personnelle Encadrés) pour les concours d'entrée aux écoles d'ingénieurs à partir de la session 2017. L'idée est de laisser LaTeX gérer la bibliographie en la complétant au passage via l'utilisation de scholar.google.fr pour les recherches et la récupération des entrée BibTeX comme décrit dans cette vidéo.