Tracking and Classification of Multiple Human Objects Directly in Compressive Measurement Domain for Low Quality Optical Videos.
Chiman KwanDavid GribbenTrac D. TranPublished in: UEMCON (2019)
Keyphrases
- low quality
- multiple objects
- high quality
- video images
- human activities
- classification accuracy
- particle filter
- d objects
- static images
- feature extraction
- dynamic scenes
- partial occlusion
- spatio temporal
- video surveillance
- video sequences
- machine learning
- image classification
- feature space
- video frames
- feature set
- palmprint
- moving camera
- image segmentation
- moving objects
- poor quality
- image processing