To search and localize objects of interest among massive and multi-modality geospatial data is key in spatial computing. However, the effective and efficient searching among an extensive collection of geospatial data (e.g., global satellite imagery, building footprint) for interesting patterns can be challenging. In this context, not only does one need to know where to look to find objects of interest but also which model to use for different searching tasks. What if prior efforts had already created models on an exact or very similar task? How should users search for such models? When models are available, how should they be stored? Many applications become possible if we manage to make large data collections and models searchable by content, metadata, and analytic tasks. Application users would like to solve these challenges by knowing which model to use, which task the model is relevant for, and how to simultaneously search across all geospatial data representations (vector, raster, text, fields, point clouds, etc.), and finding all objects of a certain type in a huge data cube (e.g., a large point cloud or time-series EO data). In the long term, users will want to search broadly, interactively, quickly, and using different or even mixed modalities.
Proceeding Downloads
Making Archives Searchable: Vision-Language Models for Classification of Historical Aerial Imagery
Historical aerial imagery archives contain valuable geospatial data for studying urban development, environmental changes, and historical events. However, the volume of data and inconsistencies in metadata and georeferencing complicate content ...
Automatic Search of Multiword Place Names on Historical Maps
Historical maps are invaluable sources of information about the past, and scanned historical maps are increasingly accessible in online libraries. To retrieve maps from these large libraries that contain specific places of interest, previous work has ...
Robust Interpolation of Arbitrary-Dimensional Moving Regions in Databases
Moving Objects Databases are designed to store, process and efficiently search for database objects with attributes that can change over time. Examples are moving points, that change position over time, or, as a more complex data type, moving regions ...
Random Affine Transformation Feature Representation Learning for Fast Polygon Retrieval
Despite the advance of representation learning (RL) of image and text data, it is a challenging task to obtain a general-purpose representation of vector-based spatial data (e.g., point, polyline, and polygon) that fulfills domain-specific prerequisites ...
Enabling Semantic-Rich Location Search on Street View Imagery Using Multilingual POI Data
Street view imagery is a valuable resource for understanding the physical environment and public health. Open-access street view imagery platforms offer street view imagery along with metadata, such as geographic coordinates and capture details (e.g., ...
Machine Learning Model Specification for Cataloging Spatio-Temporal Models (Demo Paper)
- Francis Charette-Migneault,
- Ryan Avery,
- Brian Pondi,
- Joses Omojola,
- Simone Vaccari,
- Parham Membari,
- Devis Peressutti,
- Jia Yu,
- Jed Sundwall
The Machine Learning Model (MLM) extension is a specification that extends the SpatioTemporal Asset Catalogs (STAC) framework to catalog machine learning models. This demo paper introduces the goals of the MLM, highlighting its role in improving ...
SuP-SLiP: Subsampled Processing of Large-scale Static LIDAR Point Clouds
Annotation is a crucial component of point cloud analysis. However, due to the sheer number of points in large-scale static point clouds, it is an expensive and time-consuming process. We address this issue using a novel lightweight approach to reduce ...
Index Terms
- Proceedings of the 3rd ACM SIGSPATIAL International Workshop on Searching and Mining Large Collections of Geospatial Data