Generative Adversarial Graph Convolutional Networks for Human Action Synthesis

Synthesising the spatial and temporal dynamics of the human body skeleton remains a challenging task, not only in terms of the quality of the generated shapes, but also of their diversity, particularly to synthesise realistic body movements of a specific action (action conditioning). In this paper,...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings / IEEE Workshop on Applications of Computer Vision s. 2753 - 2762
Hlavní autoři: Degardin, Bruno, Neves, Joao, Lopes, Vasco, Brito, Joao, Yaghoubi, Ehsan, Proenca, Hugo
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.01.2022
Témata:
Computer architecture
Computer vision
Convolutional codes
Deep Learning -> Graph Neural Networks Biometrics -> Human Motion Analysis/Capture; Deep Learning; Deep Learning -> Adversarial Learning; Adversarial Attack and Defense Methods; Deep Learning -> Neural Generative Models; Autoencoders; GANs
Generative adversarial networks
Measurement
Shape
Stochastic processes
ISSN:2642-9381
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Synthesising the spatial and temporal dynamics of the human body skeleton remains a challenging task, not only in terms of the quality of the generated shapes, but also of their diversity, particularly to synthesise realistic body movements of a specific action (action conditioning). In this paper, we propose Kinetic-GAN, a novel architecture that leverages the benefits of Generative Adversarial Networks and Graph Convolutional Networks to synthesise the kinetics of the human body. The proposed adversarial architecture can condition up to 120 different actions over local and global body movements while improving sample quality and diversity through latent space disentanglement and stochastic variations. Our experiments were carried out in three well-known datasets, where Kinetic-GAN notably surpasses the state-of-the-art methods in terms of distribution quality metrics while having the ability to synthesise more than one order of magnitude regarding the number of different actions. Our code and models are publicly available at https://github.com/DegardinBruno/Kinetic-GAN.
ISSN:2642-9381
DOI:10.1109/WACV51458.2022.00281