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
his project is part of the FlowNet initiative.
FlowNet aims at providing Internet freedom and free flow
information through socially informed, censor resistant online
social networks. My contribution for FLowNet is in devel-
oping an Android application, SecurePost. The requirement
for SecurePost is two-fold. First, the system should facilitate
secure, anonymous, group communication within a closed
group of trusted members. Second, the general public on the
Internet viewing this content, should be able to verify that the
content was generated only by the said closed group of trusted
members. The system consists of an Android client application,
a proxy server and a browser-plugin. The OSNs supported by
this system are Twitter and Facebook.
This is an enhanced IEEE conference template for the IEEE Conference on Sustainable Technologies. It is of primary interest for students in the "Intro to Engineering courses" (AT TÆK1002 and T-102-VERK) at Reykjavik University.