Nowak, TomaszSkrzypczyński, Piotr2023-09-252023-09-252023Nowak T., Skrzypczyński P., A New Approach to Learning of 3D Characteristic Points for Vehicle Pose Estimation. W: Progress in Polish Artificial Intelligence Research 4, Wojciechowski A. (Ed.), Lipiński P. (Ed.)., Seria: Monografie Politechniki Łódzkiej Nr. 2437, Wydawnictwo Politechniki Łódzkiej, Łódź 2023, s. 389-394, ISBN 978-83-66741-92-8, doi: 10.34658/9788366741928.61.978-83-66741-92-8http://hdl.handle.net/11652/4837https://doi.org/10.34658/9788366741928.61This article discusses the challenges of estimating the pose of a vehicle from monocular images in an uncontrolled environment. We propose a new neural network architecture that learns 3D characteristic points of vehicles from image crops and coordinates of 2D keypoints on images. To facilitate supervised training of this network, we pre-process the ApolloCar3D dataset to obtain labelled 3D characteristic points of different car models. We evaluate our approach on the ApolloCar3D benchmark and demonstrate results competitive to state-of-the-art methods.enDla wszystkich w zakresie dozwolonego użytkuFair use conditionvehicle pose estimation3D scene understandingdeep learningszacowanie pozycji pojazduzrozumienie sceny 3Dgłębokie uczenie sięRozdział - monografiaLicencja PŁLUT License10.34658/9788366741928.61