FRED is a large-scale multimodal dataset designed for drone detection, tracking, and trajectory forecasting. The dataset, authored by GabrieleMagrini, provides spatiotemporally synchronized RGB and event data. It was last updated on Hugging Face on October 3, 2025.
Use Cases
- Training drone detection models based on synchronized RGB and event data.
- Developing object tracking algorithms using the dataset's spatiotemporal sequences.
- Researching trajectory forecasting methods for drone flight paths.
- Benchmarking multimodal sensor fusion approaches for aerial object recognition.
Strengths
- Dataset is described as large-scale.
- Provides synchronized RGB and event data streams.
- Includes train and test splits, along with an alternative challenging split.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
- Source
- Hugging Face repository by GabrieleMagrini.
- Collection Method
- Likely collected using drone-mounted RGB and event cameras.
- Time Range
- null
- Freshness
- Last updated 2025-10-03 07:51:36; freshness should be verified.
- Geography
- null