Javier Barandiaran
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Javier Barandiaran has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Marcos Nieto
Vicomtech
6 shared publications - Oihana Otaegui
Vicomtech
6 shared publications - Andoni Cortes
Vicomtech-IK4
5 shared publications - Aitor Moreno
Vicomtech Foundation, Basque Research and Technology Alliance (BRTA), Mikeletegi 57, 20009 San Sebastián, Spain
2 shared publications John Congote
Vicomtech, Research Center, San Sebastian - Spain
2 shared publications- Luis Unzueta
Vicomtech
2 shared publications
Latest Publications
- Open Access
- 20 November 2026
- Published by Springer Nature
- p. 1207-1214
RailVision: An Integrated Solution for Train Detection, Speed Estimation, and Direction Analysis in Railway Environments
Accurate train detection, speed, and direction estimation are vital for railway safety and intelligent transport. This paper proposes a vision-based system integrating object detection, tracking, and homography-based motion estimation using fixed cameras. By applying geometric transformations and temporal analysis without additional hardware, the system achieves average errors below 10 km/h for speed and 10 $$^{\circ }$$ for direction. The approach maintains real-time performance with latency under 40 ms. Experimental results validate the system’s reliability, scalability, and cost-effectiveness for real-world deployment on existing infrastructure.- 20 November 2026
- Published by Springer Nature
- p. 1207-1214
Open AccessRailVision: An Integrated Solution for Train Detection, Speed Estimation, and Direction Analysis in Railway Environments
Accurate train detection, speed, and direction estimation are vital for railway safety and intelligent transport. This paper proposes a vision-based system integrating object detection, tracking, and homography-based motion estimation using fixed cameras. By applying geometric transformations and temporal analysis without additional hardware, the system achieves average errors below 10 km/h for speed and 10 $$^{\circ }$$ for direction. The approach maintains real-time performance with latency under 40 ms. Experimental results validate the system’s reliability, scalability, and cost-effectiveness for real-world deployment on existing infrastructure. - 23 October 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 23 (4) , 3115-3125
Automated Annotation of Lane Markings Using LIDAR and Odometry
Lane markings are a key element for Autonomous Driving. The generation of high definition maps and ground-truth data require extensive manual labor. In this paper, we present an efficient and robust method for the offline annotation of lane markings, using low-density LIDAR point clouds and odometry information. The odometry is used to accumulate the scans and to process them using blocks following the trajectory of the vehicle. At each block, candidate lane marking points are detected by generating virtual scan-lines and applying a dynamically optimized filter function to the LIDAR intensity values. The lane markings are tracked block wise, and their width is estimated and classified as either solid or dashed. The results are lists of connected 3D points that represent the different lane markings. The accuracy of the proposed method was tested against manually labeled recordings. A novel evaluation methodology focused on the lateral precision of detections is presented. Moreover, a web user interface was used to load the produced annotations, achieving a reduction of 60% in the annotation time, as compared to a fully manual baseline.
- 23 October 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 23 (4) , 3115-3125
Automated Annotation of Lane Markings Using LIDAR and Odometry
Lane markings are a key element for Autonomous Driving. The generation of high definition maps and ground-truth data require extensive manual labor. In this paper, we present an efficient and robust method for the offline annotation of lane markings, using low-density LIDAR point clouds and odometry information. The odometry is used to accumulate the scans and to process them using blocks following the trajectory of the vehicle. At each block, candidate lane marking points are detected by generating virtual scan-lines and applying a dynamically optimized filter function to the LIDAR intensity values. The lane markings are tracked block wise, and their width is estimated and classified as either solid or dashed. The results are lists of connected 3D points that represent the different lane markings. The accuracy of the proposed method was tested against manually labeled recordings. A novel evaluation methodology focused on the lateral precision of detections is presented. Moreover, a web user interface was used to load the produced annotations, achieving a reduction of 60% in the annotation time, as compared to a fully manual baseline.
- 28 July 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 23 (1) , 190-199
Analysis of Classifier Training on Synthetic Data for Cross-Domain Datasets
A major challenges of deep learning (DL) is the necessity to collect huge amounts of training data. Often, the lack of a sufficiently large dataset discourages the use of DL in certain applications. Typically, acquiring the required amounts of data costs considerable time, material and effort. To mitigate this problem, the use of synthetic images combined with real data is a popular approach, widely adopted in the scientific community to effectively train various detectors. In this study, we examined the potential of synthetic data-based training in the field of intelligent transportation systems. Our focus is on camera-based traffic sign recognition applications for advanced driver assistance systems and autonomous driving. The proposed augmentation pipeline of synthetic datasets includes novel augmentation processes such as structured shadows and gaussian specular highlights. A well-known DL model was trained with different datasets to compare the performance of synthetic and real image-based trained models. Additionally, a new, detailed method to objectively compare these models is proposed. Synthetic images are generated using a semi-supervised errors-guide method which is also described. Our experiments showed that a synthetic image-based approach outperforms in most cases real image-based training when applied to cross-domain test datasets (+10% precision for GTSRB dataset) and consequently, the generalization of the model is improved decreasing the cost of acquiring images.
- 28 July 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 23 (1) , 190-199
Analysis of Classifier Training on Synthetic Data for Cross-Domain Datasets
A major challenges of deep learning (DL) is the necessity to collect huge amounts of training data. Often, the lack of a sufficiently large dataset discourages the use of DL in certain applications. Typically, acquiring the required amounts of data costs considerable time, material and effort. To mitigate this problem, the use of synthetic images combined with real data is a popular approach, widely adopted in the scientific community to effectively train various detectors. In this study, we examined the potential of synthetic data-based training in the field of intelligent transportation systems. Our focus is on camera-based traffic sign recognition applications for advanced driver assistance systems and autonomous driving. The proposed augmentation pipeline of synthetic datasets includes novel augmentation processes such as structured shadows and gaussian specular highlights. A well-known DL model was trained with different datasets to compare the performance of synthetic and real image-based trained models. Additionally, a new, detailed method to objectively compare these models is proposed. Synthetic images are generated using a semi-supervised errors-guide method which is also described. Our experiments showed that a synthetic image-based approach outperforms in most cases real image-based training when applied to cross-domain test datasets (+10% precision for GTSRB dataset) and consequently, the generalization of the model is improved decreasing the cost of acquiring images.
- Open Access
- 26 July 2019
- Published by Association for Computing Machinery (ACM)
- p. 1-9
Web-based Video-Assisted Point Cloud Annotation for ADAS validation
- 26 July 2019
- Published by Association for Computing Machinery (ACM)
- p. 1-9
Open AccessWeb-based Video-Assisted Point Cloud Annotation for ADAS validation
- 8 February 2017
- Published by Springer Nature in International Journal on Interactive Design and Manufacturing (IJIDeM)
- Vol. 12 (1) , 95-103
Improved virtual reality perception with calibrated stereo and variable focus for industrial use
- 8 February 2017
- Published by Springer Nature in International Journal on Interactive Design and Manufacturing (IJIDeM)
- Vol. 12 (1) , 95-103
Improved virtual reality perception with calibrated stereo and variable focus for industrial use
- 1 May 2014
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 10 (4) , 759-769
A temporally consistent grid-based visual odometry framework for multi-core architectures
- 1 May 2014
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 10 (4) , 759-769
A temporally consistent grid-based visual odometry framework for multi-core architectures
- 1 January 2013
- Published by Springer Nature
- p. 326-339
Single Camera Railways Track Profile Inspection Using an Slice Sampling-Based Particle Filter
- 1 January 2013
- Published by Springer Nature
- p. 326-339
Single Camera Railways Track Profile Inspection Using an Slice Sampling-Based Particle Filter
- 25 November 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 13 (2) , 527-540
Adaptive Multicue Background Subtraction for Robust Vehicle Counting and Classification
In this paper, we present a robust vision-based system for vehicle tracking and classification devised for traffic flow surveillance. The system performs in real time, achieving good results, even in challenging situations, such as with moving casted shadows on sunny days, headlight reflections on the road, rainy days, and traffic jams, using only a single standard camera. We propose a robust adaptive multicue segmentation strategy that detects foreground pixels corresponding to moving and stopped vehicles, even with noisy images due to compression. First, the approach adaptively thresholds a combination of luminance and chromaticity disparity maps between the learned background and the current frame. It then adds extra features derived from gradient differences to improve the segmentation of dark vehicles with casted shadows and removes headlight reflections on the road. The segmentation is further used by a two-step tracking approach, which combines the simplicity of a linear 2-D Kalman filter and the complexity of a 3-D volume estimation using Markov chain Monte Carlo (MCMC) methods. Experimental results show that our method can count and classify vehicles in real time with a high level of performance under different environmental situations comparable with those of inductive loop detectors.
- 25 November 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 13 (2) , 527-540
Adaptive Multicue Background Subtraction for Robust Vehicle Counting and Classification
In this paper, we present a robust vision-based system for vehicle tracking and classification devised for traffic flow surveillance. The system performs in real time, achieving good results, even in challenging situations, such as with moving casted shadows on sunny days, headlight reflections on the road, rainy days, and traffic jams, using only a single standard camera. We propose a robust adaptive multicue segmentation strategy that detects foreground pixels corresponding to moving and stopped vehicles, even with noisy images due to compression. First, the approach adaptively thresholds a combination of luminance and chromaticity disparity maps between the learned background and the current frame. It then adds extra features derived from gradient differences to improve the segmentation of dark vehicles with casted shadows and removes headlight reflections on the road. The segmentation is further used by a two-step tracking approach, which combines the simplicity of a linear 2-D Kalman filter and the complexity of a 3-D volume estimation using Markov chain Monte Carlo (MCMC) methods. Experimental results show that our method can count and classify vehicles in real time with a high level of performance under different environmental situations comparable with those of inductive loop detectors.
- Open Access
- 27 October 2011
- Published by Springer Nature in EURASIP Journal on Advances in Signal Processing
- Vol. 2011 (1) , 95
Vehicle tracking and classification in challenging scenarios via slice sampling
- 27 October 2011
- Published by Springer Nature in EURASIP Journal on Advances in Signal Processing
- Vol. 2011 (1) , 95
Open AccessVehicle tracking and classification in challenging scenarios via slice sampling
- 1 June 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21612021,p. 1-4
Real-time depth map generation architecture for 3D videoconferencing
In this paper we present a reliable depth estimation system which works in real-time with commodity hardware. The system is specially intended for 3D visualization using autostereoscopic displays. The core of this work is an implementation of a modified version of the adaptive support-weight algorithm that includes highly optimized algorithms for GPU, allowing accurate and stable depth map generation. Our approach overcomes typical problems of live depth estimation systems such as depth noise and flickering. Proposed approach is integrated within the versatile GStreamer multimedia software platform. Accurate depth map estimation together with real-time performance make proposed approach suitable for 3D videoconferencing.
- 1 June 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21612021,p. 1-4
Real-time depth map generation architecture for 3D videoconferencing
In this paper we present a reliable depth estimation system which works in real-time with commodity hardware. The system is specially intended for 3D visualization using autostereoscopic displays. The core of this work is an implementation of a modified version of the adaptive support-weight algorithm that includes highly optimized algorithms for GPU, allowing accurate and stable depth map generation. Our approach overcomes typical problems of live depth estimation systems such as depth noise and flickering. Proposed approach is integrated within the versatile GStreamer multimedia software platform. Accurate depth map estimation together with real-time performance make proposed approach suitable for 3D videoconferencing.