MRFTrans: Multimodal Representation Fusion Transformer for monocular 3D semantic scene completion.
Rongtao XuJiguang ZhangJiaxi SunChangwei WangYifan WuShibiao XuWeiliang MengXiaopeng ZhangPublished in: Inf. Fusion (2024)
Keyphrases
- image sequences
- three dimensional
- semantic representation
- spatial relations
- relative depth
- fuzzy logic
- conceptual graphs
- intermediate representations
- stereo camera
- single image
- data fusion
- computer vision
- natural language
- logical representation
- input image
- multi modal
- pose estimation
- relative position
- ego motion
- scene representation
- depth cues
- semantic context
- multimodal fusion
- view invariant
- optical flow
- low level
- d scene
- vision system