Multisensor Online Transfer Learning for 3D LiDAR-based Human Detection with a Mobile Robot
This addresses the challenge of robust human detection for service robots by leveraging multisensor data, though it is incremental as it builds on existing sensor fusion and transfer learning methods.
The paper tackles the problem of improving 3D LiDAR-based human detection for mobile robots by introducing an online transfer learning framework that uses detections from other sensors (e.g., RGB-D camera, 2D LiDAR) to train the classifier via trajectory probability, resulting in performance gains as more sensors are utilized.
Human detection and tracking is an essential task for service robots, where the combined use of multiple sensors has potential advantages that are yet to be exploited. In this paper, we introduce a framework allowing a robot to learn a new 3D LiDAR-based human classifier from other sensors over time, taking advantage of a multisensor tracking system. The main innovation is the use of different detectors for existing sensors (i.e. RGB-D camera, 2D LiDAR) to train, online, a new 3D LiDAR-based human classifier, exploiting a so-called trajectory probability. Our framework uses this probability to check whether new detections belongs to a human trajectory, estimated by different sensors and/or detectors, and to learn a human classifier in a semi-supervised fashion. The framework has been implemented and tested on a real-world dataset collected by a mobile robot. We present experiments illustrating that our system is able to effectively learn from different sensors and from the environment, and that the performance of the 3D LiDAR-based human classification improves with the number of sensors/detectors used.