Discovering Latent Clusters from Geotagged Beach Images

Wang, Yan; Cao, Liangliang

doi:10.1007/978-3-642-35728-2_13

Cited by 7 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Far fewer approaches have attempted geo-localization in natural, non-urban environments. Some have limited the problem to only very specific natural environments, such as beaches [3,52], deserts [48], or mountains [1,35]. Cross-view approaches: The challenge of large-scale image geo-localization with few landmarks and limited training data has led some researchers to propose cross-view approaches, which match a query ground-level RGB image with a large reference dataset of aerial or satellite images [32,45,66].…”

Section: Single-image Geo-localizationmentioning

confidence: 99%

Where in the World is this Image? Transformer-based Geo-localization in the Wild

Pramanick¹,

Nowara²,

Gleason³

et al. 2022

Preprint

View full text Add to dashboard Cite

Predicting the geographic location (geo-localization) from a single ground-level RGB image taken anywhere in the world is a very challenging problem. The challenges include huge diversity of images due to different environmental scenarios, drastic variation in the appearance of the same location depending on the time of the day, weather, season, and more importantly, the prediction is made from a single image possibly having only a few geo-locating cues. For these reasons, most existing works are restricted to specific cities, imagery, or worldwide landmarks. In this work, we focus on developing an efficient solution to planet-scale single-image geo-localization. To this end, we propose TransLocator, a unified dual-branch transformer network that attends to tiny details over the entire image and produces robust feature representation under extreme appearance variations. TransLocator takes an RGB image and its semantic segmentation map as inputs, interacts between its two parallel branches after each transformer layer and simultaneously performs geo-localization and scene recognition in a multi-task fashion. We evaluate TransLocator on four benchmark datasets -Im2GPS [13], Im2GPS3k [14], YFCC4k [50], YFCC26k [43] and obtain 5.5%, 14.1%, 4.9%, 9.9% continent-level accuracy improvement over the state-of-the-art. TransLocator is also validated on real-world test images and found to be more effective than previous methods.

show abstract

Section: Single-image Geo-localizationmentioning

confidence: 99%

Where in the World is this Image? Transformer-based Geo-localization in the Wild

Pramanick¹,

Nowara²,

Gleason³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…En el contexto geográfico, las Máquinas de Soporte Vectorial (SVM) han sido utilizadas; siendo una de las principales técnicas de aprendizaje automático. Wang et al (2013), propuso utilizar un modelo de inferencia basado en SVM, junto con un algoritmo de agrupamiento para inferir la localización de nuevas imágenes. En Wu et al (2004) se muestran las predicciones de tiempo de traslado de un punto a otro.…”

Section: Estado Del Arteunclassified

Monitoreo urbano de entidades y eventos geográficos basado en censado social

Carrillo¹,

Ruiz²,

Moreno-Ibarra³

2019

rcar

View full text Add to dashboard Cite

La proliferación de las redes sociales y la facilidad que otorgan a sus usuarios para compartir información alienta a compartir lo que ocurre a su alrededor. Las redes sociales ayudan a conocer diferentes acontecimientos, y al estar conscientes de ellos, nos ayuda a tomar decisiones con mayor certeza. Por ejemplo, acerca del tránsito vehicular, el utilizar las redes sociales nos ayuda a saber cuándo están presentes manifestaciones o cuando ha ocurrido un accidente, esta información nos ayuda a eludir un congestionamiento vial desafortunado. Aprovechando la generación de contenido en Twitter y tomando como caso de estudio la Ciudad de México, se recolectó información de usuarios dedicados a publicar eventos viales. Por tanto, se propone una metodología para la geocodificación de textos cortos y un método de aprendizaje automático basado en Máquinas de Soporte Vectorial, con el cual se obtiene un modelo capaz de realizar un análisis espacio temporal de eventos viales. Como caso de estudio se consideró a la Ciudad de México.

show abstract