Why MCA? Nonlinear sparse coding with spike-and-slab prior for neurally plausible image encoding.
Jacquelyn A. SheltonPhilip SterneJörg BornscheinAbdul-Saboor SheikhJörg LückePublished in: NIPS (2012)
Keyphrases
- sparse coding
- image classification
- image representation
- sparse representation
- dictionary learning
- image features
- image data
- input image
- natural images
- single image super resolution
- test images
- spatial pyramid matching
- image retrieval
- multiscale
- prior model
- codebook generation
- high resolution
- visual recognition
- image segmentation
- image regions
- spatial information
- visual words
- prior knowledge
- sparse codes
- unsupervised learning
- image patches
- nearest neighbor
- object recognition
- pattern recognition
- reinforcement learning
- feature extraction