A5036291480- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Image and Video Stabilization
- Industrial Vision Systems and Defect Detection
- Image Retrieval and Classification Techniques
- Smart Parking Systems Research
- Image and Object Detection Techniques
- Cell Image Analysis Techniques
- Vehicle License Plate Recognition
- Video Coding and Compression Technologies
- Advanced Neural Network Applications
- Optical measurement and interference techniques
Universidad Autónoma de Madrid
2020-2024
Institut national de recherche en informatique et en automatique
2003
Institut de Recherche en Informatique et Systèmes Aléatoires
2003
To solve real-life problems for different smart city applications, using deep Neural Network, such as parking occupancy detection, requires fine-tuning of these networks. For large parking, it is desirable to use a cenital-plane camera located at high distance that allows the monitoring entire space or area with only one camera. Today's most popular object detection models, YOLO, achieve good precision scores real-time speed. However, if we our own data from general-purpose datasets, COCO...
Dimension reduction aims to project a high‐dimensional dataset into low‐dimensional space. It tries preserve the topological relationships among original data points and/or induce clusters. NetDRm, an online dimensionality method based on neural ensemble learning that integrates different dimension methods in synergistic way, is introduced. NetDRm designed for datasets of multidimensional can be either structured (e.g., images) or unstructured point clouds, tabular data). starts by training...
Estimating a depth map and, at the same time, predicting 3D pose of an object from single 2D color image is very challenging task. Depth estimation typically performed through stereo vision by following several time-consuming stages, such as epipolar geometry, rectification and matching. Alternatively, when not useful or applicable, relations can be inferred studied in this paper. More precisely, deep learning applied order to solve problem estimating image. Then, that used for main depicted...
This article proposes a new method which allows the detection of variations orientation video object due to rotations around axis parallel image plane. For that purpose, sequence is decomposed into temporal segments separated by key instants. The motion between these instants are therefore estimated using planar model. validity this model tested comparing their texture and shape. It assumed considered rigid has been previously segmented for each instant with high quality level. If variation...
In this paper an object-based non-planar rotation estimation for video analysis and coding is presented. This method based on a model which assumes that the moving object as planar surface. It also assumed considered objects are rigid have been previously segmented each key instant considered. The can be easily adjusted by choosing suitable region of support (block or arbitrary region). presented results indicate proposed technique approach motion estimation, used applications such MPEG-4...