Demos

Real Time Semantic Segmentation using Efficient Networks

An Efficient Network for real time semantic segmentation (even in embedded platforms). A join work with Eduardo Romera and Luis M. Bergasa.

Related Publications

Romera, Alvarez, Bergasa and Arroyo. Efficient ConvNets for Real-Time Semantic Segmentation. IV'2017.

 

Large-Scale Scene Classification in Embedded Platforms (NVIDIA TX-1)

A compact ConvNet running at real-time (65fps +) in an embedded platform. A join work with Lars Andersson and Lars Petersson.

Related Publications

José M. Álvarez and Lars Petersson. DecomposeMe: Simplifying ConvNets for end to end Learning. In Arxiv.

 

Unsupervised Image Transformation for Outdoor Semantic Labelling
Unsupervised Image Transform for Outdoor Semantic Labelling

Related Publications

German Ros and José M. Álvarez. Unsupervised Image Transformation for Outdoor Semantic Labelling. In IEEE-IV 2015.

 

Large-Scale Semantic Co-Labeling of Image Sets MSRC21

Related Publications

José M. Álvarez, Mathieu Salzmann and Nick Barnes. Large-Scale Semantic Co-Labeling of Image Sets. In WACV'14.

 

 

Road Scene Understanding from a Single Image

El contenido de esta página requiere una versión más reciente de Adobe Flash Player.

Obtener Adobe Flash Player

A Convolutional Neural network is trained using a large dataset with machine-generated labels (weak ground truth). These labels are generated using a general algorithm. The algorithm shows how convolutional neural networks can generalize from these weakly annotated data.

Related Publications

Jose M. Alvarez, Theo Gevers, Yann LeCun and Antonio M. Lopez, Road Scene Understanding from a Single Image. European Conference Computer Vision (ECCV 2012)

 

 

Combining Priors, Color and Context for Road Detection

José M. Álvarez, Antonio M. López, Theo Gevers and Felipe Lumbreras.

Related Publications

José M. Álvarez, Antonio M. Lopez, Theo Gevers and Felipe Lumbreras. Combining Priors, Color and Context for Road Detection. IEEE Trans. Intelligent Transportation Systems (ITS), to appear.

 

José M. Álvarez, Felipe Lumbreras, Theo Gevers and Antonio M. López. Geographic Information for Vision-Based Road Detection. IEEE Intelligent Vehicle Symposium (IV'10), San Diego, CA. 2010.

 

Method for Obtaining Drivable Road Area. José M. Álvarez and Felipe Lumbreras.

 

 

Road Detection Based on Illuminant Invariance

Jose M. Alvarez and Antonio M. Lopez

El contenido de esta página requiere una versión más reciente de Adobe Flash Player.

Obtener Adobe Flash Player

Road detection results rainy day Road detection results sunny day

By using an on-board camera it is possible to detect the free road surface ahead of the ego-vehicle. Road detection is of high relevance for autonomous driving, road departure warning and to support driver assistance systems such as vehicle and pedestrian detection. The key for vision-based road detection is the ability to classify image pixels as belonging or not to the road surface. Identifying road pixels is a major challenge due to the intra-class variability caused by lighting conditions. An especially difficult scenario appears when the road surface has both shadowed and non--shadowed areas. Accordingly, we propose a novel approach to vision-based road detection that is robust to shadows. The novelty of our approach relies on using a shadow invariant feature space combined with a model--based classifier The model is built on-line to improve the adaptability of the algorithm to the current lighting and the presence of other vehicles in the scene. The proposed algorithm works in still images and does not depend either on road shape or temporal restrictions. Quantitative and qualitative results are presented to support the validity of our proposal.

Code and the Road Database with manually annotated ground truth is available in the database section.

Related Publications

Jose M. Alvarez and Antonio M. Lopez, Road Detection based on Illuminant Invariance. IEEE Trans. Intelligent Transportation Systems (ITS).

Jose M. Alvarez and Antonio M. Lopez and R. Baldrich, Illuminant Invariant Model-Based Road Segmentation. IEEE Intelligent Vehicles Symposium (IV'08).

 

 

Learning Photometric Invariants for Road Detection

José M. Álvarez, Theo Gevers and Antonio M. López.

Learning photometric (illuminant) invariance Road detection

Road detection results

Related Publications

José M. Álvarez, Theo Gevers and Antonio M. López. Learning Photometric Invariance for Object Detection. In International Journal of Computer Vision Volume 90, Number 1, 45-6, 2010.

José M. Álvarez, Theo Gevers and Antonio M. Lopez. Learning Photometric Invariance from Diversified Color Model Ensembles. In Proc. IEEE Int. Conf. on Computer Vision and Pattern Recognition (CVPR'09), Miami, Fl 2009.

 

Computer-Aided System for Morphometric Mandibular Index Computation (using dental panoramic radiographs)

José M. Álvarez and Jose López-López.

 

 

Visit the webpage of the project.

Related Publications

"J. López-López, Jose M. Alvarez, E. Jané-Salas, A. Estrugo-Devesa, R. Ayuso-Montero, E. Velasco-Ortega, Computer-Aided System for Morphometric Mandibular Index Computation (using dental panoramic radiographs). Dentistry, Oral Surgery & Medicine

 

Dynamic Comparison of Headlights

J. Serrat, F. Diego, F. Lumbreras, José M. Álvarez and Antonio M. López.

 

 

Visit the webpage of the project.

Related Publications

J. Serrat, F. Diego, F. Lumbreras, José M. Álvarez, Antonio M. López and C. Elvira. Dynamic comparison of headlights. In Jourmal of Automobile Engineering, Proc. of the Institution of Mechanical Engineers, Part D, June 2007

 

tumblr site counter