Diverse Motion In-betweening with Dual Posture Stitching
Not on the map. Why?
Abstract (arXiv)
In-betweening is a technique for generating transitions given initial and target character states. The majority of existing works require multiple (often >10) frames as input, which are not always accessible. Our work deals with a focused yet challenging problem: to generate the transition when given exactly two frames (only the first and last). To cope with this challenging scenario, we implement our bi-directional scheme which generates forward and backward transitions from the start and end frames with two adversarial autoregressive networks, and stitches them in the middle of the transition where there is no strict ground truth. The autoregressive networks based on conditional variational autoencoders (CVAE) are optimized by searching for a pair of optimal latent codes that minimize a novel stitching loss between their outputs. Results show that our method achieves higher motion quality and more diverse results than existing methods on both the LaFAN1 and Human3.6m datasets.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 26 of its 45 references that are in the map. First posted on arXiv 25 Mar 2023.
Builds on
26 of 45 references.
References · 26
- 2016Least Squares Generative Adversarial Networkscited by 1,000
- 2021Action-Conditioned 3D Human Motion Synthesis with Transformer VAEcited by 497
- 2017On human motion prediction using recurrent neural networkscited by 336
- 2015Structural-RNN: Deep Learning on Spatio-Temporal Graphscited by 234
- 2021Robust Motion In-betweeningcited by 223
- 2020Long-term Human Motion Prediction with Scene Contextcited by 191
- 2020We are More than Our Joints: Predicting how 3D Bodies Movecited by 115
- 2018Human Motion Prediction via Spatio-Temporal Inpaintingcited by 114
- 2018QuaterNet: A Quaternion-based Recurrent Model for Human Motioncited by 108
- 2019Structured Prediction Helps 3D Human Motion Modellingcited by 96
- 2017The Pose Knows: Video Forecasting by Generating Pose Futurescited by 88
- 2018A Neural Temporal Model for Human Motion Predictioncited by 87
14 more
- 2020Convolutional Autoencoders for Human Motion Infillingcited by 73
- 2018BiHMP-GAN: Bidirectional 3D Human Motion Prediction GANcited by 55
- 2018Recurrent Transition Networks for Character Locomotioncited by 52
- 2017Deep Video Generation, Prediction and Completion of Human Action Sequencescited by 50
- 2018Human Motion Modeling using DVGANscited by 45
- 2021LoBSTr: Real-time Lower-body Pose Prediction from Sparse Upper-body Tracking Signalscited by 45
- 2018Combining Recurrent Neural Networks and Adversarial Training for Human Motion Synthesis and Controlcited by 39
- 2022Real-time Controllable Motion Transition for Characterscited by 29
- 2022Conditional Motion In-betweeningcited by 25
- 2020Hierarchical Style-based Networks for Motion Synthesiscited by 23
- 2019PVRED: A Position-Velocity Recurrent Encoder-Decoder for Human Motion Predictioncited by 20
- 2020Probabilistic Character Motion Synthesis using a Hierarchical Deep Latent Variable Modelcited by 16
- 2021TrajeVAE: Controllable Human Motion Generation from Trajectoriescited by 14
- 2019Human Motion Prediction via Pattern Completion in Latent Representation Spacecited by 2
Built on by
Citing papers per year: 2023: 1; 2024: 1.
2 papers.