FE-Fusion-VPR: Attention-based Multi-Scale Network Architecture for Visual Place Recognition by Fusing Frames and Events.
Kuanxu HouDelei KongJunjie JiangHao ZhuangXinjie HuangZheng FangPublished in: CoRR (2022)
Keyphrases
- network architecture
- multiscale
- artificial neural
- fusion method
- selective attention
- neural network
- event detection
- visual information
- neural network model
- image fusion
- fusing multiple
- feature fusion
- temporal order
- data fusion
- activation function
- information fusion
- finite element
- service integration
- natural images
- scale space
- connection weights
- combining multiple
- video frames
- visual processing
- interesting events
- decision level
- visual field
- coarse to fine
- visual attention
- visual features
- edge detection
- network infrastructure
- soft tissue
- fusion process
- multi sensor
- video clips
- temporal information