Rastreamento de pedestres 3D multi-câmera usando redes neurais de grafos

Abstract

Tracking the position of pedestrians over time through camera images is a rising computer vision research topic. In multi-camera settings, the researches are even more recent. Many solutions use supervised neural networks to solve this problem, which can require a lot of effort to annotate the data in addition to a lot of time spent to train the network. The goals of this work are: develop variations of pedestrian tracking algorithms, being desirable to avoid the need to have annotated data; and compare the results obtained through accuracy metrics. Therefore, this work proposes an approach for tracking pedestrians in 3D space in multi-camera environments using the Message Passing Neural Network framework inspired by graphs. We evaluated the solution using the WILDTRACK dataset and a generalizable detection method, reaching 77.1% of MOTA when training with data obtained by a generalizable tracking algorithm. The algorithm can track at a 40 frames per second rate.