Stacked dense optical flows and dropout layers to predict sperm motility and morphology.
Vajira ThambawitaPål HalvorsenHugo HammerMichael RieglerTrine B. HaugenPublished in: CoRR (2019)
Keyphrases
- optical flow
- motion parameters
- motion estimation
- image processing
- scene flow
- video endoscopy
- flow estimation
- multi layer
- camera motion
- image sequences
- gray scale
- missing data
- data sets
- motion model
- point correspondences
- motion segmentation
- model selection
- image analysis
- moving camera
- morphological operators
- morphological analysis
- moving objects
- neural network