SIVED: SAR Vehicle Detection with Oriented Bounding Boxes
Available on 1 platform
Sign in to view source links and access this dataset
Description
A Synthetic Aperture Radar (SAR) image dataset for vehicle detection. The dataset includes oriented bounding box annotations, which are a specialized format for object localization. It is hosted on Kaggle, but details on the author, organization, size, and update date are not provided.
Use Cases
Train object detection models for vehicles in SAR imagery.
Benchmark the performance of detection algorithms on radar-based images.
Develop models for oriented object detection, which can handle rotated targets.
Research domain adaptation techniques for applying models from optical to SAR data.
Strengths
Includes oriented bounding box annotations, which provide more precise localization than standard boxes.
Focuses on Synthetic Aperture Radar (SAR) imagery, a less common but important data modality.
Limitations
Dataset size, row count, and file formats are unknown, limiting suitability assessment.
Column-level documentation and sample data are absent; field semantics must be inferred after download.
Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
Source
Kaggle
Collection Method
Method of collection is not described.
Time Range
Temporal coverage is not specified.
Freshness
Last update date is unknown; freshness unverified.
Geography
Spatial coverage is not specified.
License information is unknown; users must verify licensing terms before use.