DSC: Dense-Sparse Convolution for Vectorized Inference of Convolutional Neural Networks.
Alexander FrickensteinManoj Rohit VemparalaChristian UngerFatih AyarWalter StechelePublished in: CVPR Workshops (2019)
Keyphrases
- convolutional neural networks
- convolutional network
- dense stereo
- dense optical flow
- image processing
- motion field estimation
- sparse data
- belief networks
- inference process
- high dimensional
- probabilistic inference
- bayesian inference
- dense motion estimation
- knowledge base
- compressive sensing
- fast fourier transform
- fir filters
- fourier transform
- image reconstruction
- sparse representation
- signal processing