SG-NN: Sparse Generative Neural Networks for Self-Supervised Scene Completion of RGB-D Scans.
Angela DaiChristian DillerMatthias NießnerPublished in: CVPR (2020)
Keyphrases
- neural network
- camera tracking
- indoor scenes
- back propagation
- three dimensional
- d scene
- artificial neural networks
- laser scanner
- depth images
- neural network model
- video sequences
- feed forward
- generative model
- optic flow field
- pattern recognition
- fuzzy logic
- genetic algorithm
- single image
- multi layer
- fault diagnosis
- image sequences
- modular neural networks
- backpropagation neural network
- dynamic scenes
- neural nets
- training process
- nearest neighbor
- activation function
- input image
- scene understanding
- moving objects
- high dimensional
- feed forward neural networks
- stereo camera
- multi layer perceptron
- training algorithm
- recurrent neural networks
- multilayer perceptron
- fuzzy inference system
- bp neural network
- neyman pearson
- pose estimation