Automatic Human Pose Annotation for Loose-fitting Clothes

Matsumoto, Takuya; Shimosato, Kodai; Maeda, Takahiro; Murakami, Tatsuya; Murakoso, Koji; Mino, Kazuhiko; Ukita, Norimichi

doi:10.23919/mva.2019.8757927

Cited by 4 publications

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Human motion prediction, which forecasts future body poses based on past poses, is a crucial technique for human-robot interaction [22,23,28,29,41], autonomous driving [30], VR/AR applications [18], performance capture [8,32,33,43,44], etc. However, these applications are limited because of the lack of training motion data, which results in low prediction accuracy.…”

Section: Introductionmentioning

confidence: 99%

Data Augmentation for Human Motion Prediction

Maeda

Ukita

2021

2021 17th International Conference on Machine Vision and Applications (MVA)

Self Cite

View full text Add to dashboard Cite

This paper presents a motion data augmentation scheme incorporating motion synthesis encouraging diversity and motion correction imposing physical plausibility. This motion synthesis consists of our modified Variational AutoEncoder (VAE) and Inverse Kinematics (IK). In this VAE, our proposed sampling-nearsamples method generates various valid motions even with insufficient training motion data. Our IK-based motion synthesis method allows us to generate a variety of motions semi-automatically. Since these two schemes generate unrealistic artifacts in the synthesized motions, our motion correction rectifies them. This motion correction scheme consists of imitation learning with physics simulation and subsequent motion debiasing. For this imitation learning, we propose the PD-residual force that significantly accelerates the training process. Furthermore, our motion debiasing successfully offsets the motion bias induced by imitation learning to maximize the effect of augmentation. As a result, our method outperforms previous noise-based motion augmentation methods by a large margin on both Recurrent Neural Network-based and Graph Convolutional Network-based human motion prediction models. The code is available at https: //github.com/meaten/MotionAug.

show abstract