Polyp Tracking in Video Colonoscopy Using Optical Flow With an On-The-Fly Trained CNN.
He ZhengHanbo ChenJunzhou HuangXuzhi LiXiao HanJianhua YaoPublished in: ISBI (2019)
Keyphrases
- optical flow
- moving camera
- feature tracking
- motion model
- dense optical flow
- lucas kanade
- face detection and tracking
- object segmentation and tracking
- motion segmentation
- video sequences
- single frame
- image sequences
- video images
- object detection and tracking
- person detection
- surveillance videos
- video surveillance
- real time
- video data
- articulated human motion
- moving objects
- point tracking
- stationary camera
- camera motion
- image frames
- motion features
- video streams
- motion field
- motion estimation
- optical flow computation
- video frames
- cellular neural networks
- d scene
- object tracking
- space time
- video dataset
- multimedia
- object motion
- dynamic scenes
- eye tracking
- optical flow estimation
- motion analysis
- high frame rate
- computer vision
- pre trained
- flow field
- particle filter
- dynamic textures
- video shots
- colorectal cancer
- spatio temporal
- training set
- appearance model
- structure from motion
- video clips
- temporal continuity
- motion parameters
- video retrieval
- video content
- kalman filter
- neural network