Ivan Pino - Academia.edu (original) (raw)
Uploads
Papers by Ivan Pino
2017 European Conference on Mobile Robots (ECMR), 2017
Vehicle detection and tracking is a core ingredient for developing autonomous driving application... more Vehicle detection and tracking is a core ingredient for developing autonomous driving applications in urban scenarios. Recent image-based Deep Learning (DL) techniques are obtaining breakthrough results in these perceptive tasks. However, DL research has not yet advanced much towards processing 3D point clouds from lidar range-finders. These sensors are very common in autonomous vehicles since, despite not providing as semantically rich information as images, their performance is more robust under harsh weather conditions than vision sensors. In this paper we present a full vehicle detection and tracking system that works with 3D lidar information only. Our detection step uses a Convolutional Neural Network (CNN) that receives as input a featured representation of the 3D information provided by a Velodyne HDL-64 sensor and returns a per-point classification of whether it belongs to a vehicle or not. The classified point cloud is then geometrically processed to generate observations for a multi-object tracking system implemented via a number of Multi-Hypothesis Extended Kalman Filters (MH-EKF) that estimate the position and velocity of the surrounding vehicles. The system is thoroughly evaluated on the KITTI tracking dataset, and we show the performance boost provided by our CNN-based vehicle detector over a standard geometric approach. Our lidar-based approach uses about a 4% of the data needed for an image-based detector with similarly competitive results.
IEEE Transactions on Intelligent Transportation Systems, 2020
We present a novel vehicle detection and tracking system that works solely on 3D LiDAR informatio... more We present a novel vehicle detection and tracking system that works solely on 3D LiDAR information. Our approach segments vehicles using a dual-view representation of the 3D LiDAR point cloud on two independently trained convolutional neural networks, one for each view. A bounding box growing algorithm is applied to the fused output of the networks to properly enclose the segmented vehicles. Bounding boxes are grown using a probabilistic method that takes into account also occluded areas. The final vehicle bounding boxes act as observations for a multi-hypothesis tracking system which allows to estimate the position and velocity of the observed vehicles. We thoroughly evaluate our system on the KITTI benchmarks both for detection and tracking separately and show that our dualbranch classifier consistently outperforms previous single-branch approaches, improving or directly competing to other state of the art LiDAR-based methods. Index Terms-Deep convolutional neural network, vehicle detection and tracking, LiDAR, point cloud.
En el marco del proyecto ARREL se ha experimentado con diversos sistemas de captura masiva de dat... more En el marco del proyecto ARREL se ha experimentado con diversos sistemas de captura masiva de datos con el objetivo de documentar el circo romano y su actual entorno urbano. Ello ha permitido construir el escenario de un serious game pero, al mismo tiempo ha representado una oportunidad para crear un portal público de consulta y visualización de la realidad arquitectónica de este recinto histórico. Con ello se avanza en el desarrollo y experimentación de “plataformas a la carta” donde el usuario puede avanzar en el conocimiento técnico de una realidad patrimonial diacrónica y entender el espacio urbano como la etapa final de un dilatado proceso histórico.In the framework of Project ‘ARREL’ several Massive Data Capture Systems (MDCS) have been tried, with the goal of documenting the Roman circus and its current urban context. This has enabled us to construct the stage of a serious game but, at the same time, we have been able to set up an online site of the architectural reality of this historical precinct, that can be publicly accessed for visualization and consultation. With it, we advance in the development and experimentation of “menu platforms”, where the user can freely choose from the available technical knowledge in a vast diachronical heritage reality, and ultimately understand urban space as the final state of an expanded historical process
2017 European Conference on Mobile Robots (ECMR), 2017
Vehicle detection and tracking is a core ingredient for developing autonomous driving application... more Vehicle detection and tracking is a core ingredient for developing autonomous driving applications in urban scenarios. Recent image-based Deep Learning (DL) techniques are obtaining breakthrough results in these perceptive tasks. However, DL research has not yet advanced much towards processing 3D point clouds from lidar range-finders. These sensors are very common in autonomous vehicles since, despite not providing as semantically rich information as images, their performance is more robust under harsh weather conditions than vision sensors. In this paper we present a full vehicle detection and tracking system that works with 3D lidar information only. Our detection step uses a Convolutional Neural Network (CNN) that receives as input a featured representation of the 3D information provided by a Velodyne HDL-64 sensor and returns a per-point classification of whether it belongs to a vehicle or not. The classified point cloud is then geometrically processed to generate observations for a multi-object tracking system implemented via a number of Multi-Hypothesis Extended Kalman Filters (MH-EKF) that estimate the position and velocity of the surrounding vehicles. The system is thoroughly evaluated on the KITTI tracking dataset, and we show the performance boost provided by our CNN-based vehicle detector over a standard geometric approach. Our lidar-based approach uses about a 4% of the data needed for an image-based detector with similarly competitive results.
IEEE Transactions on Intelligent Transportation Systems, 2020
We present a novel vehicle detection and tracking system that works solely on 3D LiDAR informatio... more We present a novel vehicle detection and tracking system that works solely on 3D LiDAR information. Our approach segments vehicles using a dual-view representation of the 3D LiDAR point cloud on two independently trained convolutional neural networks, one for each view. A bounding box growing algorithm is applied to the fused output of the networks to properly enclose the segmented vehicles. Bounding boxes are grown using a probabilistic method that takes into account also occluded areas. The final vehicle bounding boxes act as observations for a multi-hypothesis tracking system which allows to estimate the position and velocity of the observed vehicles. We thoroughly evaluate our system on the KITTI benchmarks both for detection and tracking separately and show that our dualbranch classifier consistently outperforms previous single-branch approaches, improving or directly competing to other state of the art LiDAR-based methods. Index Terms-Deep convolutional neural network, vehicle detection and tracking, LiDAR, point cloud.
En el marco del proyecto ARREL se ha experimentado con diversos sistemas de captura masiva de dat... more En el marco del proyecto ARREL se ha experimentado con diversos sistemas de captura masiva de datos con el objetivo de documentar el circo romano y su actual entorno urbano. Ello ha permitido construir el escenario de un serious game pero, al mismo tiempo ha representado una oportunidad para crear un portal público de consulta y visualización de la realidad arquitectónica de este recinto histórico. Con ello se avanza en el desarrollo y experimentación de “plataformas a la carta” donde el usuario puede avanzar en el conocimiento técnico de una realidad patrimonial diacrónica y entender el espacio urbano como la etapa final de un dilatado proceso histórico.In the framework of Project ‘ARREL’ several Massive Data Capture Systems (MDCS) have been tried, with the goal of documenting the Roman circus and its current urban context. This has enabled us to construct the stage of a serious game but, at the same time, we have been able to set up an online site of the architectural reality of this historical precinct, that can be publicly accessed for visualization and consultation. With it, we advance in the development and experimentation of “menu platforms”, where the user can freely choose from the available technical knowledge in a vast diachronical heritage reality, and ultimately understand urban space as the final state of an expanded historical process