Jorge Hernández

Point cloud segmentation towards urban ground modeling

This paper presents a new method for segmentation and interpretation of 3D point clouds from mobile LIDAR data. The main contribution of this work is the automatic detection and classification of artifacts located at the ground level. The detection is based on Top-Hat of hole filling algorithm of range images. Then, several features are extracted from the detected connected components (CCs). Afterward, a stepwise forward variable selection by using Wilks Lambda criterion is performed. Finally, CCs are classified in four categories (lampposts, pedestrians, cars, the others) by using a SVM machine learning method.

Morphological segmentation of building façade images

In this paper, we describe an automatic method for segmentation of building facade images. First, individual facades are isolated from general city block images. This step is based on accumulation of directional color gradients, assuming that facade structures are aligned. Then sky region is detected based on segmentation approach and color marker extraction. Finally, the images are split in floors using directional color gradient accumulation, as well. Our approach introduces several morphological filters to augment the robustness to problems such as: textured balconies, some specular reflections of the bright windows and small obstacles in images. The experimental results show the performance of our approach.

Shape ultimate attribute opening

The ultimate opening (UO) is a powerful segmentation operator recently introduced by Beucher [1]. It automatically selects the most contrasted regions of an image. However, in the presence of nested structures (e.g. text in a signboard or windows in a contrasted facade), interesting structures may be masked by the containing region. In this paper we focus on ultimate attribute openings and we propose a method that improves the results by favoring regions with a predefined shape via a similarity function. An efficient implementation using a max-tree representation of the image is proposed. The method is validated in the framework of three applications: facade analysis, scene-text detection and cell segmentation. Experimental results show that the proposed method yields better segmentation results than UO.

Segmentation of Façades from Urban 3D Point Clouds Using Geometrical and Morphological Attribute-Based Operators

3D building segmentation is an important research issue in the remote sensing community with relevant applications to urban modeling, cloud-to-cloud and cloud-to-model registration, 3D cartography, virtual reality, cultural heritage documentation, among others. In this paper, we propose automatic, parametric and robust approaches to segment facades from 3D point clouds. Processing is carried out using elevation images and 3D decomposition, and the final result can be reprojected onto the 3D point cloud for visualization or evaluation purposes. Our methods are based on geometrical and geodesic constraints. Parameters are related to urban and architectural constraints. Thus, they can be set up to manage facades of any height, length and elongation. We propose two methods based on facade marker extraction and a third method without markers based on the maximal elongation image. This work is developed in the framework of TerraMobilita project. The performance of our methods is proved in our experiments on TerraMobilita databases using 2D and 3D ground truth annotations.

Ultimate Attribute Opening Segmentation with Shape Information

In this paper, a method for morphological segmentation using shape information is presented. This method is based on a morphological operator named ultimate attribute opening (UAO ). Our approach considers shape information to favor the detection of specific shapes. The method is validated in the framework of two applications: facade analysis and scene-text detection. The experimental results show that our approach is more robust than the standard UAO .

Ultimate Opening Combined with Area Stability Applied to Urban Scenes

This paper explores the use of ultimate opening in urban analysis context. It demonstrates the efficiency of this approach for street level elevation images, derived from 3D point clouds acquired by terrestrial mobile mapping systems. An area-stability term is introduced in the residual definition, reducing the over-segmentation of the vegetation while preserving small significant regions.

Quality assessment of monocular 3D inference

Recently proliferation of 3D inference methods shows an important alternative to perceive in 3D of real world from single images. The quality evaluation of 3D estimated from inference methods has been demonstrated using dataset with 3D ground truth data. However in real scenarios, the 3D inference quality is complete unknown. In this work, we present a new quality assessment of 3D monocular inference. First, we define the notion of quality index for 3D inference data. Then, we present a weighted linear model of similarity metrics to estimate quality index. The method is based on hand crafted similarity measures among image representations of RGB image and 3D inferred data. We demonstrate the effectiveness of our proposed method using public datasets and 3D inference methods of state of the art.