Block-Matching Convolutional Neural Network (BMCNN): Improving CNN-Based Denoising by Block-Matched Inputs.
Byeongyong AhnYoonsik KimGu Yong ParkNam Ik ChoPublished in: APSIPA (2018)
Keyphrases
- convolutional neural network
- block matching
- denoising
- motion estimation
- face detection
- optical flow
- motion compensation
- motion vectors
- video compression
- block size
- block matching algorithm
- hierarchical block matching
- natural images
- image processing
- neural network
- frequency domain
- object detection
- motion compensated
- blocking artifacts
- prediction error
- computer vision
- reference frame
- video coding
- motion field
- feature space
- image sequences
- multimedia