Now showing items 1-3 of 3
Regression-Based Locating Landmark on Dynamic Humans
We present a novel framework that consists of two-level regressors for finding correlations between human shapes and landmark positions in both body part and holistic scales. To this end, we first develop pose invariant ...
Motion Retargeting to Preserve Spatial Relationship between Skinned Characters
Applying motion capture data for multi-person interaction to virtual characters is challenging because one needs to preserve interaction semantics in addition to satisfying the general requirements for motion retargeting, ...
Latent Motion Manifold with Sequential Auto-Encoders
(The Eurographics Association, 2018)
We propose the sequential autoencoders for constructing latent motion manifold. Sequential autoencoders minimize the difference between the ground truth motion space distribution and reconstructed motion space distribution ...