Hybrid bayesian convolutional neural network object detection architectures for tracking small markers in automotive crashtest videos.
Felix NeubürgerDaniel GierseThomas KopinskiPublished in: NLDL (2023)
Keyphrases
- convolutional neural network
- object detection
- face detection
- human detection
- real time
- video surveillance
- bayesian networks
- moving camera
- surveillance videos
- scene understanding
- appearance model
- background subtraction
- neural network
- video images
- bayesian estimation
- posterior probability
- object tracking
- computer vision
- kalman filter
- sequential monte carlo
- bayesian formulation
- moving vehicles
- object recognition
- multi class
- maximum likelihood
- particle filter
- moving target
- pedestrian detection
- particle filtering
- mean shift
- object segmentation
- video scene
- video dataset
- successive frames
- visual tracking
- video stabilization
- foreground background segmentation
- machine learning