GaussianFormer: Scene as Gaussians for Vision-Based 3D Semantic Occupancy Prediction.
Yuanhui HuangWenzhao ZhengYunpeng ZhangJie ZhouJiwen LuPublished in: CoRR (2024)
Keyphrases
- prediction accuracy
- d scene
- real time
- semantic web
- semantic context
- human computer interaction
- video sequences
- semantic information
- prediction error
- multiple images
- vision sensors
- dynamic scenes
- linear combination
- single image
- prediction model
- scene analysis
- semantic network
- road scenes
- semantic annotation
- moving objects
- augmented reality
- vision system
- domain specific
- complex scenes
- natural language
- image sequences
- low level features
- gaussian mixture model
- mixture model
- scene understanding
- real scenes
- outdoor scenes
- prediction algorithm
- object recognition