DINOSTAR: Deep Iterative Neural Object Detector Self-Supervised Training for Roadside LiDAR Applications

TL;DR

This paper introduces a self-supervised framework for training deep object detectors on roadside LiDAR data, eliminating the need for human annotations while achieving comparable performance to supervised methods.

Contribution

The novel self-supervised training framework uses statistically modeled teachers to generate noisy labels, enabling scalable and annotation-free training of deep detectors for roadside point-cloud data.

Findings

Achieves comparable performance to human-annotated training.

Reduces time and cost associated with manual labeling.

Enhances detector robustness through diverse point-cloud representations.

Abstract

Recent advancements in deep-learning methods for object detection in point-cloud data have enabled numerous roadside applications, fostering improvements in transportation safety and management. However, the intricate nature of point-cloud data poses significant challenges for human-supervised labeling, resulting in substantial expenditures of time and capital. This paper addresses the issue by developing an end-to-end, scalable, and self-supervised framework for training deep object detectors tailored for roadside point-cloud data. The proposed framework leverages self-supervised, statistically modeled teachers to train off-the-shelf deep object detectors, thus circumventing the need for human supervision. The teacher models follow fine-tuned set standard practices of background filtering, object clustering, bounding-box fitting, and classification to generate noisy labels. It is presented that by training the student model over the combined noisy annotations from multitude of teachers enhances its capacity to discern background/foreground more effectively and forces it to learn diverse point-cloud-representations for object categories of interest. The evaluations, involving publicly available roadside datasets and state-of-art deep object detectors, demonstrate that the proposed framework achieves comparable performance to deep object detectors trained on human-annotated labels, despite not utilizing such human-annotations in its training process.