TLCFuse: Temporal Multi-Modality Fusion Towards Occlusion-Aware Semantic Segmentation.
Gustavo Salazar-GomezWenqian LiuManuel Diaz-ZapataDavid Sierra GonzálezChristian LaugierPublished in: IV (2024)
Keyphrases
- multi modality
- semantic segmentation
- street scenes
- multi modal
- information theoretic
- medical images
- image registration
- conditional random fields
- scene classification
- mutual information
- superpixels
- object categories
- imaging modalities
- weakly supervised
- object class
- object classes
- medical imaging
- image classification
- object detection
- image analysis
- pairwise
- partial occlusion
- bounding box
- image intensity
- pascal voc
- information extraction
- optical flow