Single-Shot 3D Multi-Person Shape Reconstruction from a Single RGB Image
Although the performance of the 3D human shape reconstruction method has improved considerably in recent years, most methods focus on a single person, reconstruct a root-relative 3D shape, and rely on ground-truth information about the absolute depth to convert the reconstruction result to the camer...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-07-01
|
Series: | Entropy |
Subjects: | |
Online Access: | https://www.mdpi.com/1099-4300/22/8/806 |
id |
doaj-a931d38b085b4e5a9bd14ad5356b62be |
---|---|
record_format |
Article |
spelling |
doaj-a931d38b085b4e5a9bd14ad5356b62be2020-11-25T03:26:33ZengMDPI AGEntropy1099-43002020-07-012280680610.3390/e22080806Single-Shot 3D Multi-Person Shape Reconstruction from a Single RGB ImageSeong Hyun Kim0Ju Yong Chang1Department of Electronics and Communication Engineering, Kwangwoon University, Seoul 01897, KoreaDepartment of Electronics and Communication Engineering, Kwangwoon University, Seoul 01897, KoreaAlthough the performance of the 3D human shape reconstruction method has improved considerably in recent years, most methods focus on a single person, reconstruct a root-relative 3D shape, and rely on ground-truth information about the absolute depth to convert the reconstruction result to the camera coordinate system. In this paper, we propose an end-to-end learning-based model for single-shot, 3D, multi-person shape reconstruction in the camera coordinate system from a single RGB image. Our network produces output tensors divided into grid cells to reconstruct the 3D shapes of multiple persons in a single-shot manner, where each grid cell contains information about the subject. Moreover, our network predicts the absolute position of the root joint while reconstructing the root-relative 3D shape, which enables reconstructing the 3D shapes of multiple persons in the camera coordinate system. The proposed network can be learned in an end-to-end manner and process images at about 37 fps to perform the 3D multi-person shape reconstruction task in real time.https://www.mdpi.com/1099-4300/22/8/8063D human shape reconstructionstatistical body shape modeldeep neural network |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Seong Hyun Kim Ju Yong Chang |
spellingShingle |
Seong Hyun Kim Ju Yong Chang Single-Shot 3D Multi-Person Shape Reconstruction from a Single RGB Image Entropy 3D human shape reconstruction statistical body shape model deep neural network |
author_facet |
Seong Hyun Kim Ju Yong Chang |
author_sort |
Seong Hyun Kim |
title |
Single-Shot 3D Multi-Person Shape Reconstruction from a Single RGB Image |
title_short |
Single-Shot 3D Multi-Person Shape Reconstruction from a Single RGB Image |
title_full |
Single-Shot 3D Multi-Person Shape Reconstruction from a Single RGB Image |
title_fullStr |
Single-Shot 3D Multi-Person Shape Reconstruction from a Single RGB Image |
title_full_unstemmed |
Single-Shot 3D Multi-Person Shape Reconstruction from a Single RGB Image |
title_sort |
single-shot 3d multi-person shape reconstruction from a single rgb image |
publisher |
MDPI AG |
series |
Entropy |
issn |
1099-4300 |
publishDate |
2020-07-01 |
description |
Although the performance of the 3D human shape reconstruction method has improved considerably in recent years, most methods focus on a single person, reconstruct a root-relative 3D shape, and rely on ground-truth information about the absolute depth to convert the reconstruction result to the camera coordinate system. In this paper, we propose an end-to-end learning-based model for single-shot, 3D, multi-person shape reconstruction in the camera coordinate system from a single RGB image. Our network produces output tensors divided into grid cells to reconstruct the 3D shapes of multiple persons in a single-shot manner, where each grid cell contains information about the subject. Moreover, our network predicts the absolute position of the root joint while reconstructing the root-relative 3D shape, which enables reconstructing the 3D shapes of multiple persons in the camera coordinate system. The proposed network can be learned in an end-to-end manner and process images at about 37 fps to perform the 3D multi-person shape reconstruction task in real time. |
topic |
3D human shape reconstruction statistical body shape model deep neural network |
url |
https://www.mdpi.com/1099-4300/22/8/806 |
work_keys_str_mv |
AT seonghyunkim singleshot3dmultipersonshapereconstructionfromasinglergbimage AT juyongchang singleshot3dmultipersonshapereconstructionfromasinglergbimage |
_version_ |
1724592136407482368 |