SYNTHIA, The  SYNTHetic collection of Imagery and Annotations, is a dataset that has been generated with the purpose of aiding semantic segmentation and related scene understanding problems in the context of driving scenarios. SYNTHIA consists of a collection of photo-realistic frames rendered from a virtual city and comes with precise pixel-level semantic annotations for 13 classes: misc, sky, building, road, sidewalk, fence, vegetation, pole, car, sign, pedestrian, cyclist, lanemarking.




  • Large volume of data & groundtruth: +200,000 HD images from video streams and +20,000 HD images from independent snapshots

  • Scene diversity: European style town, modern city, highway and green areas

  • Variety of dynamic objects: cars, pedestrians and cyclists

  • Multiple seasons: dedicated themes for winter, fall, spring and summer

  • Lighting conditions and weather: dynamic lights and shadows, several day-time modes, rain mode and night mode

  • Sensor simulation: 8 RGB cameras forming a binocular 360º camera, 8 depth sensors

  • Automatic groundtruth: individual instances for semantic segmentation (pixelwise annotations), depth, car ego-motion


Ready to start using SYNTHIA?

Download it!


Our CVPR 2016 paper [pdf ] describes our virtual environment along with our novel proposal for domain adaptation of semantic segmentation models based on state-of-the-art CNNs.