BEVWorld: A Multimodal World Model for Autonomous Driving via Unified BEV Latent Space.
Yumeng ZhangShi GongKaixin XiongXiaoqing YeXiao TanFan WangJizhou HuangHua WuHaifeng WangPublished in: CoRR (2024)
Keyphrases
- world model
- autonomous driving
- latent space
- grand challenge
- latent variables
- stereo vision
- low dimensional
- gaussian process
- vision system
- dimensionality reduction
- lower dimensional
- generative model
- manifold learning
- semantic interpretation
- parameter space
- feature space
- matrix factorization
- gaussian processes
- probabilistic latent semantic analysis
- high dimensional
- transfer learning
- distance metric
- urban traffic
- gaussian process latent variable models
- vision algorithms
- high dimensional data
- pattern recognition
- gaussian mixture
- original data
- random variables
- labeled data
- principal component analysis
- probabilistic model
- prior knowledge
- data analysis