Object-based fusion for urban tree species classification from hyperspectral,panchromatic and nDSM data

This study aims at identifying the best object-based fusion strategy that takes advantage of the complementarity of several heterogeneous airborne data sources for improving the classification of 15 tree species in an urban area (Toulouse, France). The airborne data sources are: hyperspectral Visible Near-Infrared (160 spectral bands, spatial resolution of 0.4 m) and Short-Wavelength Infrared (256 spectral bands, 1.6 m), panchromatic (14 cm), and a normalized Digital Surface Model (12.5 cm). Object-based feature and decision level fusion strategies are proposed and compared when applied to a reference site where the species are previously identified during ground truth collection. This allows the best fusion strategy to be selected with a view to introducing the method in an automatic process (tree crown delineation and species classification) on a test site, independent of the reference site used for learning. In particular, a decision level fusion is selected: based on the Support Vector Machine algorithm, Visible Near-Infrared and Short-Wavelength Infrared classifications use Minimum Noise Fraction components at the original spatial resolution, whereas panchromatic and normalized Digital Surface Model classifications use, respectively, Haralick’s and structural features computed at the object scale. After the computation of a decision profile for each source at the object level based on the classification algorithms’ membership probabilities, these decision profiles are combined and a decision rule is applied to predict the species. Focusing on the reference site, the Visible Near-Infrared exhibits the best performances with F-score values higher than 60% for 13 species out of 15. The Short-Wavelength Infrared is the most powerful for three species with F-score greater than 60% for seven common species with the Visible Near-Infrared. The panchromatic and normalized Digital Surface Model contribute marginally. The best fusion strategy (decision fusion) does not improve significantly the overall accuracy with 77% (kappa = 74%) against 75% (kappa = 72%) for the Visible Near-Infrared but in general, it improves the results for cases where complementarities have been observed. When applied to the test site and assessed for the two majority species (Tilia tomentosa and Platanus x hispanica), the selected approach gives consistent results with an overall accuracy of 63% against 55% for the Visible Near-Infrared.     

Using immersive virtual reality and anatomically correct computer-generated characters in the forensic assessment of deviant sexual preferences

Penile plethysmography (PPG) is the gold standard for the assessment of sexual interests, especially among sex offenders of children. Nonetheless, this method faces some ethical limitations inherent to the nature of its stimuli and could benefit from the improvement of its ecological validity. The use of computer-generated characters (CGC) in virtual immersion for PPG assessment might help address these issues. A new application developed to design made-to-measure anatomically correct virtual characters compatible with the Tanner developmental stages is presented. The main purpose of this study was to determine how the virtual reality (VR) modality compares to the standard auditory modality on their capacity to generate sexual arousal profiles and deviance differentials indicative of sexual interests. The erectile responses of 22 sex offenders of children and 42 non-deviant adult males were recorded. While both stimulus modalities generated significantly different genital arousal profiles for sex offenders of children and non-deviant males, deviance differentials calculated from the VR modality allowed for significantly higher classification accuracy. Performing receiver operating characteristic analyses further assessed discriminant potential. Auditory modality yielded an area under the curve (AUC) of 0.79 (SE = 0.059) while CGC in VR yielded an AUC of 0.90 (SE = 0.052). Overall, results suggest that the VR modality allows significantly better group classification accuracy and discriminant validity than audio stimuli, which provide empirical support for the use of this new method for PPG assessment. Additionally, the potential use of VR in interventions pertaining to self-regulation of sexual offending is addressed in conclusion.     

Leveraging Blur Information for Plenoptic Camera Calibration

International Journal of Computer Vision - This paper presents a novel calibration algorithm for plenoptic cameras, especially the multi-focus configuration, where several types of micro-lenses are...     

An enumeration algorithm for solving the fleet management problem in underground mines

This paper presents an enumeration algorithm based on dynamic programming for optimally solving the fleet management problem in underground mines. This problem consists of routing and scheduling bidirectional vehicles on a haulage network composed of one-lane bidirectional road segments. The method takes into account the displacement modes of the vehicles, either forward or in reverse, and makes sure that these vehicles move forward when they arrive at their service point. The method has been developed for the underground mine context, but it can be extended to the industrial environment.     

Valence-Driven Connectivity Encoding for 3D Meshes

In this paper, we propose a valence-driven, single-resolution encoding technique for lossless compression of triangle mesh connectivity. Building upon a valence-based approach pioneered by Touma and Gotsman²² , we design a new valence-driven conquest for arbitrary meshes that always guarantees smaller compression rates than the original method. Furthermore, we provide a novel theoretical entropy study of our technique, hinting the optimality of the valence-driven approach. Finally, we demonstrate the practical efficiency of this approach (in agreement with the theoretical prediction) on a series of test meshes, resulting in the lowest compression ratios published so far, for both irregular and regular meshes, small or large.     

Exact Algorithms for <Emphasis Type="Italic">L</Emphasis>(2,1)-Labeling of Graphs

The notion of distance constrained graph labelings, motivated by the Frequency Assignment Problem, reads as follows: A mapping from the vertex set of a graph G=(V,E) into an interval of integers {0,…,k} is an L(2,1)-labeling of G of span k if any two adjacent vertices are mapped onto integers that are at least 2 apart, and every two vertices with a common neighbor are mapped onto distinct integers. It is known that for any fixed k≥4, deciding the existence of such a labeling is an NP-complete problem. We present exact exponential time algorithms that are faster than the naive O ^*((k+1) ⁿ ) algorithm that would try all possible mappings. The improvement is best seen in the first NP-complete case of k=4, where the running time of our algorithm is O(1.3006 ⁿ ). Furthermore we show that dynamic programming can be used to establish an O(3.8730 ⁿ ) algorithm to compute an optimal L(2,1)-labeling.     

RTAB‐Map as an open‐source lidar and visual simultaneous localization and mapping library for large‐scale and long‐term online operation

Distributed as an open‐source library since 2013, real‐time appearance‐based mapping (RTAB‐Map) started as an appearance‐based loop closure detection approach with memory management to deal with large‐scale and long‐term online operation. It then grew to implement simultaneous localization and mapping (SLAM) on various robots and mobile platforms. As each application brings its own set of constraints on sensors, processing capabilities, and locomotion, it raises the question of which SLAM approach is the most appropriate to use in terms of cost, accuracy, computation power, and ease of integration. Since most of SLAM approaches are either visual‐ or lidar‐based, comparison is difficult. Therefore, we decided to extend RTAB‐Map to support both visual and lidar SLAM, providing in one package a tool allowing users to implement and compare a variety of 3D and 2D solutions for a wide range of applications with different robots and sensors. This paper presents this extended version of RTAB‐Map and its use in comparing, both quantitatively and qualitatively, a large selection of popular real‐world datasets (e.g., KITTI, EuRoC, TUM RGB‐D, MIT Stata Center on PR2 robot), outlining strengths, and limitations of visual and lidar SLAM configurations from a practical perspective for autonomous navigation applications.     

Deformation Transfer to Multi‐Component Objects

We present a simple and effective algorithm to transfer deformation between surface meshes with multiple components. The algorithm automatically computes spatial relationships between components of the target object, builds correspondences between source and target, and finally transfers deformation of the source onto the target while preserving cohesion between the target's components. We demonstrate the versatility of our approach on various complex models.