Pedestrian detection at night from a RGB camera is an under-represented yet very important problem, where current state-of-the-art vision algorithms fail. Computer vision methods for detection at night have not received much attention, despite the fact they are a critical building block of many systems such as safe and robust autonomous cars.
To further assess and advance the state of the art, for the second year we organize the NightOwls Pedestrian Detetection Challenge 2020, as part of the Scalability in Autonomous Driving Workshop, CVPR 2020.
The competition uses the NightOwls dataset, consisting of 279,000 fully-annotated images in 40 video sequences recorded at night, captured by an industry-standard camera. This year, we extend the competition into three tracks:
- Pedestrian detection from a single frame
- Pedestrian detection from multiple frames
- Object detection and classification at night (pedestrians, cyclists and motorbikes)
The image data is the same for all three tracks, but each track has its own leaderboard and we will announce the winners in each track individually. Data annotations are available for the training/validation subset, but the annotations for the testing set are kept only on the evaluation server to ensure fair competition.
The winner will be announced at the Scalability in Autonomous Driving workshop, CVPR 2020 on 15th June and will be presented with valuable and unique prices, by courtesy of the Visual Geometry Group, University of Oxford 🙂
- March 2020 – training and validation sets published (images + annotations)
- 18.05.2020 – testing set published (images only), submission site opens for competition entries
- 31.05.2020 23:59 UTC – competition entries submission deadline
- 15.06.2020 – winners announced at Scalability in Autonomous Driving workshop, CVPR 2020
Lukas Neumann, Andrea Vedaldi, Andrew Zisserman, Bernt Schiele