English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
 
 
DownloadE-Mail
  Pose-Guided Human Animation from a Single Image in the Wild

Yoon, J. S., Liu, L., Golyanik, V., Sarkar, K., Park, H. S., & Theobalt, C. (2020). Pose-Guided Human Animation from a Single Image in the Wild. Retrieved from https://arxiv.org/abs/2012.03796.

Item is

Files

show Files
hide Files
:
arXiv:2012.03796.pdf (Preprint), 3MB
Name:
arXiv:2012.03796.pdf
Description:
File downloaded from arXiv at 2021-02-08 14:04
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-

Locators

show

Creators

show
hide
 Creators:
Yoon, Jae Shin1, Author
Liu, Lingjie2, Author              
Golyanik, Vladislav2, Author              
Sarkar, Kripasindhu1, Author
Park, Hyun Soo1, Author
Theobalt, Christian2, Author              
Affiliations:
1External Organizations, ou_persistent22              
2Computer Graphics, MPI for Informatics, Max Planck Society, ou_40047              

Content

show
hide
Free keywords: Computer Science, Computer Vision and Pattern Recognition, cs.CV
 Abstract: We present a new pose transfer method for synthesizing a human animation from a single image of a person controlled by a sequence of body poses. Existing pose transfer methods exhibit significant visual artifacts when applying to a novel scene, resulting in temporal inconsistency and failures in preserving the identity and textures of the person. To address these limitations, we design a compositional neural network that predicts the silhouette, garment labels, and textures. Each modular network is explicitly dedicated to a subtask that can be learned from the synthetic data. At the inference time, we utilize the trained network to produce a unified representation of appearance and its labels in UV coordinates, which remains constant across poses. The unified representation provides an incomplete yet strong guidance to generating the appearance in response to the pose change. We use the trained network to complete the appearance and render it with the background. With these strategies, we are able to synthesize human animations that can preserve the identity and appearance of the person in a temporally coherent way without any fine-tuning of the network on the testing scene. Experiments show that our method outperforms the state-of-the-arts in terms of synthesis quality, temporal coherence, and generalization ability.

Details

show
hide
Language(s): eng - English
 Dates: 2020-12-072020
 Publication Status: Published online
 Pages: 14 p.
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: arXiv: 2012.03796
URI: https://arxiv.org/abs/2012.03796
BibTex Citekey: Yoon_2012.03796
 Degree: -

Event

show

Legal Case

show

Project information

show

Source

show