This paper provides a comprehensive survey of road infrastructure perception for autonomous driving, organizing existing research into geometric-structural, functional-spatial, and regulatory-semantic categories. It reviews sensing modalities, scene representations, methods, and benchmarks, highlighting the field's shift towards more sophisticated scene modeling. The survey also identifies limitations in current benchmarks regarding reasoning, consistency, and planning-oriented assessment, and discusses future research directions for unified representation, generalization, and evaluation. AI
IMPACT Provides a foundational taxonomy and identifies key challenges for advancing autonomous driving perception systems.
RANK_REASON The item is an academic paper surveying a specific technical field within AI. [lever_c_demoted from research: ic=1 ai=1.0]
- alphaXiv
- arXiv
- autonomous driving
- CatalyzeX Code Finder for Papers
- Connected Papers
- CORE Recommender
- DagsHub
- drivable-space estimation
- Gotit.pub
- Hugging Face
- Influence Flower
- Litmaps
- online map construction
- road infrastructure perception
- ScienceCast
- scite Smart Citations
- traffic sign recognition
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →