Unlimited Road-scene Synthetic Annotation (URSA) Dataset.
Matt AngusMohamed ElBalkiniSamin KhanAli HarakehOles AndrienkoCody ReadingSteven Lake WaslanderKrzysztof CzarneckiPublished in: CoRR (2018)
Keyphrases
- real scenes
- road scenes
- d scene
- road vehicles
- video sequences
- metadata
- indoor and outdoor scenes
- semantic annotation
- image sequences
- single image
- three dimensional
- image set
- annotated images
- gps data
- benchmark datasets
- weather conditions
- weakly labeled
- outdoor images
- traffic scenes
- image retrieval
- scene understanding
- real world
- video annotation
- autonomous driving
- complex scenes
- multiple images
- depth map
- object recognition
- aerial images
- scene images
- photo collections
- dynamic scenes
- video data
- urban areas
- manual annotation
- automatic annotation
- outdoor scenes
- synthetic datasets
- image collections
- stereo vision
- moving objects
- optical flow