A5072435381- Image Enhancement Techniques
- Advanced Vision and Imaging
- Advanced Image Fusion Techniques
- Remote Sensing and LiDAR Applications
- Advanced Image Processing Techniques
- Forest ecology and management
- Forest Ecology and Biodiversity Studies
- Color Science and Applications
- Remote Sensing in Agriculture
- Image and Signal Denoising Methods
- Image and Video Quality Assessment
- Species Distribution and Climate Change
Swiss Federal Institute for Forest, Snow and Landscape Research
2025
Université de Rennes
2015-2018
Abstract Proximally sensed laser scanning presents new opportunities for automated forest ecosystem data capture. However, a gap remains in deriving ecologically pertinent information, such as tree species, without additional ground data. Artificial intelligence approaches, particularly deep learning (DL), have shown promise towards automation. Progress has been limited by the lack of large, diverse, and, most importantly, openly available labelled single‐tree point cloud datasets. This...
Proximally-sensed laser scanning offers significant potential for automated forest data capture, but challenges remain in automatically identifying tree species without additional ground data. Deep learning (DL) shows promise automation, yet progress is slowed by the lack of large, diverse, openly available labeled datasets single point clouds. This has impacted robustness DL models and ability to establish best practices classification. To overcome these challenges, FOR-species20K benchmark...
Multivariate generalized Gaussian distributions (MGGDs) have aroused a great interest in the image processing community thanks to their ability describe accurately various features, such as gradient fields. However, so far applicability has been limited by lack of transformation between two these parametric distributions. In this paper, we propose novel MGGDs, consisting an optimal transportation second-order statistics and stochastic-based shape parameter transformation. We employ proposed...
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
In this paper, we propose a perceptual model for evaluating results from color transfer methods. We conduct user study, which provides set of subjective scores triplets input, target and result images. Then, each triplet, compute number image features, objectively characterize transfer. To describe the relationship between these features scores, build regression with random forests. An analysis cross-validation show that predictions our are highly accurate.
In this paper, we propose a novel transformation between two Beta distributions. Our progressively and accurately reshapes an input distribution into target using four intermediate statistical transformations. The key idea of paper is to adopt the model discrete distributions color light in images. We design new which apply context transfer Experiments have shown that our method obtains more natural less saturated results than recent state-of-the-art methods. Moreover, portray better both...
Color transfer methods alter the look of a source image with regards to reference image. So far, proposed color have been limited low-dynamic-range (LDR) images. Unlike LDR images, which are display-dependent, high-dynamic-range (HDR) images contain real physical values world luminance and able capture high variations finest details scenes. Therefore, there exists strong discrepancy between two types In this paper, we bridge gap domain HDR imagery by introducing extensions methods. We tackle...