Diversity-Promoting Human Motion Interpolation via Conditional Variational Auto-Encoder
This addresses the problem of creating realistic and varied human animations for applications like gaming or simulation, but it is incremental as it builds on existing CVAE and RNN techniques.
The paper tackled generating diverse human motion interpolations between start and end poses using a Conditional Variational Auto-Encoder with RNNs and a regularization loss, achieving plausible and varied results as demonstrated on a public dataset.
In this paper, we present a deep generative model based method to generate diverse human motion interpolation results. We resort to the Conditional Variational Auto-Encoder (CVAE) to learn human motion conditioned on a pair of given start and end motions, by leveraging the Recurrent Neural Network (RNN) structure for both the encoder and the decoder. Additionally, we introduce a regularization loss to further promote sample diversity. Once trained, our method is able to generate multiple plausible coherent motions by repetitively sampling from the learned latent space. Experiments on the publicly available dataset demonstrate the effectiveness of our method, in terms of sample plausibility and diversity.