Gallery — Conference Paper

Paper template for ICon-MaSTEd submissions

Modelo de artigo para submissão no IX Simpósio de Iniciação Científica da Engenharia Elétrica (IX SICEEL) realizado durante a Semana de integração da Engenharia Elétrica (SIEEL), em parceria com a Universidade Federal de São Carlos (UFSCar) e Universidade de São Paulo (USP)

Template in English for SBSR (Brazilian Symposium on Remote Sensing)

This document contains instructions for preparing ACL 2016 submissions and camera-ready manuscripts. The document itself conforms to its own specifications, and is therefore an example of what your manuscript should look like. Papers are required to conform to all the directions reported in this document. By using the provided LaTeX and BibTeX styles (acl2016.sty, acl2016.bst), the required formatting will be enabled by default.

A template for creating ICES conference proceedings submissions.

Template LaTeX pour les 22èmes journées Nationales Microondes 7 au 10 juin 2022 - Limoges Date limite de soumission : 06 décembre 2021 https://jnm2022.sciencesconf.org/

Template for Submission to IJCAI-19; downloaded from the conference's Author's Kit page.

Template for the Interactive Sonification (ISon) workshop proceedings.

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.