Multiple object tracking using feature fusion in hierarchical LSTMs

Multiple object tracking using feature fusion in hierarchical LSTMs

Multiple object tracking sets the foundation for many intelligent video applications. The authors present a novel tracking solution using the ability of recurrent neural networks to effectively model complex temporal dynamics between objects irrespective of appearances, pose, occlusions, and illumin...

Full description

Bibliographic Details
Main Author: Ehtesham Hassan
Format: Article
Language:English
Published: Wiley 2020-09-01
Series:The Journal of Engineering
Subjects:
image sequences
object tracking
object detection
image representation
video signal processing
image segmentation
image motion analysis
recurrent neural nets
learning (artificial intelligence)
feature extraction
track modelling
motion coding scheme
relative position
motion representation
hierarchical lstm structure
track association
multiple object tracking challenge datasets
feature fusion
hierarchical lstm
intelligent video applications
recurrent neural networks
complex temporal dynamics
online tracking
active tracks
tracking-by-detection methodology
hierarchical long short term memory network structure
motion dynamics
motion cues
bounding boxes
object instance segments
Online Access:https://digital-library.theiet.org/content/journals/10.1049/joe.2020.0111

Internet

https://digital-library.theiet.org/content/journals/10.1049/joe.2020.0111

Similar Items