NeuralAnnot: Neural Annotator for in-the-wild Expressive 3D Human Pose and Mesh Training Sets.
Gyeongsik MoonKyoung Mu LeePublished in: CoRR (2020)
Keyphrases
- training set
- neural network
- d mesh
- network architecture
- training examples
- monocular images
- markerless
- supervised learning
- active learning
- feature space
- mesh generation
- bio inspired
- neural model
- body parts
- classification accuracy
- training data
- biologically plausible
- hebbian learning
- neural fuzzy
- class distribution
- associative memory
- semantic annotation
- high curvature
- triangle mesh
- nonlinear predictive control
- volumetric data
- motion tracking
- visual hull
- svm classifier
- decision trees
- computer vision