Evidential deep learning for arbitrary LIDAR object classification in the context of autonomous driving

Edouard Capellier (1) , Franck Davoine (2) , Véronique Cherfaoui (2) , You Li (3)

1 Technocentre Renault [Guyancourt]
2 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]
3 Key Laboratory for Information Technologies in Agriculture

In traditional LIDAR processing pipelines, a point-cloud is split into clusters, or objects, which are classified afterwards. This supposes that all the objects obtained by clustering belong to one of the classes that the classifier can recognize, which is hard to guarantee in practice. We thus propose an evidential end-to-end deep neural network to classify LIDAR objects. The system is capable of classifying ambiguous and incoherent objects as unknown, while only having been trained on vehicles and vulnerable road users. This is achieved thanks to an evidential reformulation of generalized logistic regression classifiers, and an online filtering strategy based on statistical assumptions. The training and testing were realized on LIDAR objects which were labelled in a semi-automatic fashion, and collected in different situations thanks to an autonomous driving and perception platform.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Intelligence artificielle [cs.AI]; Robotique [cs.RO]

Edouard Capellier, Franck Davoine, Véronique Cherfaoui, You Li. Evidential deep learning for arbitrary LIDAR object classification in the context of autonomous driving. 30th IEEE Intelligent Vehicles Symposium (IV 2019), Jun 2019, Paris, France. pp.1304-1311. ⟨hal-02322434⟩