Method And System For Virtual Sensor Data Generation With Depth Ground Truth Annotation

Abstract

Methods and systems for generating virtual sensor data for developing or testing computer vision detection algorithms are described. A system and a method may involve generating a virtual environment. The system and the method may also involve positioning a virtual sensor at a first location in the virtual environment. The system and the method may also involve recording data characterizing the virtual environment, the data corresponding to information generated by the virtual sensor sensing the virtual environment. The system and the method may further involves annotating the data with a depth map characterizing a spatial relationship between the virtual sensor and the virtual environment.

Description

technical field [0001] The present disclosure relates generally to virtual data generation and annotation, and more particularly to methods and systems for generating virtual sensor data suitable for developing, training, testing, and validating vehicle-based computer vision algorithms for estimating depth of objects from images . Background technique [0002] In the development of computer vision algorithms using supervised machine learning to detect objects in the driving environment, different sets of sensor data are required to train, develop, test and validate the detection algorithms and additional downstream functions associated with the algorithms. However, this often requires significant amounts of money, time, and resources to acquire real-world sensor data. For example, in order to obtain real-world sensor data, sensors may need to be physically located or mounted on the vehicle along the driving path, and actual driving operations may need to be performed on var...

AI Technical Summary

Problems solved by technology

Therefore, the number of sensor data sets to be collected can be enormous

Images