Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

LEVERAGING ORDINAL REGRESSION WITH SOFT LABELS FOR 3D HEAD POSE ESTIMATION FROM POINT SETS

Abstract: 

Head pose estimation from depth image is a challenging problem, considering its large pose variations, severer occlusions, and low quality of depth data. In contrast to existing approaches that take 2D depth image as input, we propose a novel deep regression architecture called Head PointNet, which consumes 3D point sets derived from a depth image describing the visible surface of a head. To cope with the non-stationary property of pose variation process, the network is facilitated with an ordinal regression module that incorporates metric penalties into ground truth label representation. The soft label representation encodes inter-class and intra-class information contained in the class labels simultaneously, and guides the network to learn discriminative features. Experiments on two challenging datasets, namely the Biwi Head Pose Dataset and Pandora Dataset, show that our proposed method outperforms state-of-the-art approaches.

up
0 users have voted:

Paper Details

Authors:
Shihua Xiao, Nan Sang, Xupeng Wang, Xiangtian Ma
Submitted On:
12 February 2020 - 4:44am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Xupeng Wang
Paper Code:
5344
Document Year:
2020
Cite

Document Files

presentation_paperID5344.pptx

(51)

Subscribe

[1] Shihua Xiao, Nan Sang, Xupeng Wang, Xiangtian Ma, "LEVERAGING ORDINAL REGRESSION WITH SOFT LABELS FOR 3D HEAD POSE ESTIMATION FROM POINT SETS", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/4981. Accessed: Jul. 13, 2020.
@article{4981-20,
url = {http://sigport.org/4981},
author = {Shihua Xiao; Nan Sang; Xupeng Wang; Xiangtian Ma },
publisher = {IEEE SigPort},
title = {LEVERAGING ORDINAL REGRESSION WITH SOFT LABELS FOR 3D HEAD POSE ESTIMATION FROM POINT SETS},
year = {2020} }
TY - EJOUR
T1 - LEVERAGING ORDINAL REGRESSION WITH SOFT LABELS FOR 3D HEAD POSE ESTIMATION FROM POINT SETS
AU - Shihua Xiao; Nan Sang; Xupeng Wang; Xiangtian Ma
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/4981
ER -
Shihua Xiao, Nan Sang, Xupeng Wang, Xiangtian Ma. (2020). LEVERAGING ORDINAL REGRESSION WITH SOFT LABELS FOR 3D HEAD POSE ESTIMATION FROM POINT SETS. IEEE SigPort. http://sigport.org/4981
Shihua Xiao, Nan Sang, Xupeng Wang, Xiangtian Ma, 2020. LEVERAGING ORDINAL REGRESSION WITH SOFT LABELS FOR 3D HEAD POSE ESTIMATION FROM POINT SETS. Available at: http://sigport.org/4981.
Shihua Xiao, Nan Sang, Xupeng Wang, Xiangtian Ma. (2020). "LEVERAGING ORDINAL REGRESSION WITH SOFT LABELS FOR 3D HEAD POSE ESTIMATION FROM POINT SETS." Web.
1. Shihua Xiao, Nan Sang, Xupeng Wang, Xiangtian Ma. LEVERAGING ORDINAL REGRESSION WITH SOFT LABELS FOR 3D HEAD POSE ESTIMATION FROM POINT SETS [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/4981