Action2Motion: Conditioned Generation of 3D Human Motions
arXiv:2007.15240 · doi:10.1145/3394171.3413635
Abstract
Action recognition is a relatively established task, where givenan input sequence of human motion, the goal is to predict its ac-tion category. This paper, on the other hand, considers a relativelynew problem, which could be thought of as an inverse of actionrecognition: given a prescribed action type, we aim to generateplausible human motion sequences in 3D. Importantly, the set ofgenerated motions are expected to maintain itsdiversityto be ableto explore the entire action-conditioned motion space; meanwhile,each sampled sequence faithfully resembles anaturalhuman bodyarticulation dynamics. Motivated by these objectives, we followthe physics law of human kinematics by adopting the Lie Algebratheory to represent thenaturalhuman motions; we also propose atemporal Variational Auto-Encoder (VAE) that encourages adiversesampling of the motion space. A new 3D human motion dataset, HumanAct12, is also constructed. Empirical experiments overthree distinct human motion datasets (including ours) demonstratethe effectiveness of our approach.
13 pages, ACM MultiMedia 2020
Cited by in corpus (12)
- InterGen: Diffusion-based Multi-human Motion Generation under Complex Interactions
- 4D Facial Expression Diffusion Model
- Pose-aware Attention Network for Flexible Motion Retargeting by Body Part
- Text-to-Motion Retrieval: Towards Joint Understanding of Human Motion Data and Natural Language
- A personalized time-resolved 3D mesh generative model for unveiling normal heart dynamics
- A Survey on Human Interaction Motion Generation
- Motion Diffusion Autoencoders: Enabling Attribute Manipulation in Human Motion Demonstrated on Karate Techniques
- ANT: Adaptive Neural Temporal-Aware Text-to-Motion Model
- A Standardized Benchmark for Skeleton-Based Rehabilitation Assessment Using Deep Learning
- Crafting Dynamic Virtual Activities with Advanced Multimodal Models
- Spline-based Transformers
- ARDY: Autoregressive Diffusion with Hybrid Representation for Interactive Human Motion Generation