Designed multimodal (image + radar heat map) FIFO-TransFuser backbone model (segmentation) on foggy situation
- TransFuser input: radar heat map and output feature map of FIFO model (FIFO generate image feature map that is robust to foggy situations)
- FIFO model input: image
- Backbone: backbone.py
- Inference: test.py
- .pth file path: ./FIFO_final_model.pth (.pth file can be obtained from FIFO GitHub (see below))
- FIFO: Learning Fog-invariant Features for Foggy Scene Segmentation
GitHub: FIFO - TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Driving
GitHub: TransFuser

