The reviewed record of science sign in
Pith

arxiv: 2203.07706 · v2 · pith:BWUYKR2G · submitted 2022-03-15 · cs.CV · cs.GR· cs.MM

ActFormer: A GAN-based Transformer towards General Action-Conditioned 3D Human Motion Generation

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:BWUYKR2Grecord.jsonopen to challenge →

classification cs.CV cs.GRcs.MM
keywords motionhumanmulti-persontransformergenerationactformeraction-conditionedgeneral
0
0 comments X
read the original abstract

We present a GAN-based Transformer for general action-conditioned 3D human motion generation, including not only single-person actions but also multi-person interactive actions. Our approach consists of a powerful Action-conditioned motion TransFormer (ActFormer) under a GAN training scheme, equipped with a Gaussian Process latent prior. Such a design combines the strong spatio-temporal representation capacity of Transformer, superiority in generative modeling of GAN, and inherent temporal correlations from the latent prior. Furthermore, ActFormer can be naturally extended to multi-person motions by alternately modeling temporal correlations and human interactions with Transformer encoders. To further facilitate research on multi-person motion generation, we introduce a new synthetic dataset of complex multi-person combat behaviors. Extensive experiments on NTU-13, NTU RGB+D 120, BABEL and the proposed combat dataset show that our method can adapt to various human motion representations and achieve superior performance over the state-of-the-art methods on both single-person and multi-person motion generation tasks, demonstrating a promising step towards a general human motion generator.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.