METHOD FOR GENERATING DATA SET FOR LEARNING FOR DETECTION OF OBSTACLE IN AUTONOMOUS DRIVING CIRCUMSTANCES AND COMPUTING DEVICE, LEARNING METHOD, AND LEARNING DEVICE USING THE SAME

摘要

To provide a method for generating data set for learning for detecting an obstacle in autonomous driving circumstances.SOLUTION: A computing device includes the steps of: obtaining at least one first original image indicating road driving circumstances and a first segmentation ground truth image corresponding to the first original image; obtaining a second original image including a specific object and a second segmentation ground truth image corresponding to the specific object; obtaining a third original image created by cutting a portion corresponding to the specific object and a third segmentation ground truth image created by cutting pixels corresponding to a portion where the specific object is located; and creating at least one fourth original image by compositing the third original image with the first original image and creating a data set for learning including the fourth original image and a fourth segmentation ground truth image corresponding to the fourth original image.SELECTED DRAWING: Figure 1