Neural Fields as Learnable Kernels for 3D Reconstruction.
Francis WilliamsZan GojcicSameh KhamisDenis ZorinJoan BrunaSanja FidlerOr LitanyPublished in: CoRR (2021)
Keyphrases
- computer vision
- network architecture
- neural network
- kernel methods
- learning algorithm
- kernel function
- support vector
- structure from motion
- three dimensional
- neural model
- linear combination
- positive examples
- biologically plausible
- feature space
- background knowledge
- d scene
- learning problems
- multiple kernel learning
- pac learning
- hypothesis space
- concept class
- positive definite