FIFNET: A convolutional neural network for motion-based multiframe super-resolution using fusion of interpolated frames.
Hamed ElwarfalliRussell C. HardiePublished in: Comput. Vis. Image Underst. (2021)
Keyphrases
- multi frame
- super resolution
- convolutional neural network
- low resolution
- high resolution
- single frame
- motion estimation
- multiple frames
- video super resolution
- super resolution image reconstruction
- super resolution reconstruction
- low resolution images
- resolution enhancement
- image sequences
- face detection
- optical flow
- motion analysis
- motion blur
- optical flow estimation
- high resolution images
- high quality
- point correspondences
- pairwise
- image super resolution
- space time
- optic flow
- motion field
- motion vectors
- video coding
- object recognition
- motion model
- computer vision