Shape Matters: Understanding the Implicit Bias of the Noise Covariance.
Jeff Z. HaoChenColin WeiJason D. LeeTengyu MaPublished in: CoRR (2020)
Keyphrases
- image structure
- shape model
- arbitrary shape
- noise level
- noise reduction
- shape descriptors
- imaging artifacts
- measurement error
- noise model
- additive noise
- shape analysis
- multiscale
- residual error
- median filter
- covariance matrix
- scale space
- input data
- shape features
- medial axis
- signal to noise ratio
- missing data
- noisy environments
- shape recognition
- noise free
- image edges
- medical images
- low variance
- computer vision