Figure 4From: Adaptive Multi-modal Fusion Instance Segmentation for CAEVs in Complex Conditions: Dataset, Framework and VerificationsReal complex environment and corresponding simulated conditions on the KITTI dataset. a Night, b and e Light rain day, c Light rain night, d Glare. The real environment images a–d are from the Oxford RobotCar dataset [50], e is from EuroCity Persons dataset [51]Back to article page