E
Edgar Roman-Rangel
Researcher at University of Geneva
Publications - 41
Citations - 348
Edgar Roman-Rangel is an academic researcher from University of Geneva. The author has contributed to research in topics: Computer science & Image retrieval. The author has an hindex of 8, co-authored 35 publications receiving 260 citations. Previous affiliations of Edgar Roman-Rangel include Instituto Tecnológico Autónomo de México & Idiap Research Institute.
Papers
More filters
Journal ArticleDOI
Analyzing Ancient Maya Glyph Collections with Contextual Shape Descriptors
TL;DR: The approach is promising, as it improves performance on the retrieval task, has been successfully validated under an epigraphic viewpoint, and has the potential of offering both novel insights in archeology and practical solutions for real daily scholar needs.
Journal ArticleDOI
Building Hierarchical Representations for Oracle Character and Sketch Recognition
TL;DR: A novel hierarchical representation is proposed that combines a Gabor-related low-level representation and a sparse-encoder-related mid- level representation that has beaten humans at recognizing general sketches.
Journal ArticleDOI
Fig Plant Segmentation from Aerial Images Using a Deep Convolutional Encoder-Decoder Network
Jorge Fuentes-Pacheco,Juan Torres-Olivares,Edgar Roman-Rangel,Salvador Cervantes,Porfirio Juárez-López,Jorge Hermosillo-Valadez,Juan Manuel Rendón-Mancha +6 more
TL;DR: This work presents a Convolutional Neural Network with an encoder-decoder architecture that classifies each pixel as crop or non-crop using only raw colour images as input and achieves a mean accuracy of 93.85% despite the complexity of the background and a highly variable visual appearance of the leaves.
Proceedings ArticleDOI
Searching the past: an improved shape descriptor to retrieve maya hieroglyphs
TL;DR: The introduction and analysis of a new dataset of 3400+ Maya hieroglyphs, whose compilation involved manual search, annotation and segmentation by experts, presents several challenges for visual description and automatic retrieval as it is rich in complex visual details.
Proceedings ArticleDOI
Retrieving ancient Maya glyphs with Shape Context
TL;DR: An improvement in the cost function used to compute similarity between shapes making it more restrictive and precise is proposed, and the results are promising, they are analyzed via standard image retrieval measurements.