A complementary dual-backbone transformer extracting and fusing weak cues for object detection in extremely dark videos.
Bo ZhangJinli SuoQionghai DaiPublished in: Inf. Fusion (2023)
Keyphrases
- object detection
- contextual cues
- motion cues
- scene understanding
- fuzzy logic
- video sequences
- background subtraction
- face detection
- human detection
- video database
- object segmentation
- video frames
- video data
- object categories
- video surveillance
- object recognition
- video analysis
- moving camera
- action detection
- pedestrian detection
- primal dual
- visual patterns
- object classification
- neural network
- power system
- multi class
- multimedia
- computer vision
- visual cues
- combining multiple
- video content
- human activities
- mid level
- video search
- event recognition
- high level
- high voltage