PETS2025 invites submissions for the following challenges:
Challenge 1: Target Detection and Classification (Mandatory)
Objective: Detecting and classifying targets in maritime and coastal areas using multi-platform and multi-spectral sensors
- Apply one or more object detectors to all of the provided test image sequences (if possible, for comprehensive evaluation), or at least two sequences from different sensor types.
- Report results in a standardised format.
- Include a brief description of the methodology used, with results submitted in XML format.
Challenge 2: Long-Term Target Tracking (Optional)
Objective: Tracking multiple targets across diverse terrains using multi-platform and multi-spectral sensors
- Apply one or more object trackers to all of the provided test image sequences (if possible, for comprehensive evaluation), or at least two sequences from different sensor types.
- Report results in a standardised format.
- Include a brief description of the methodology used, with results submitted in XML format.
Challenge 3: Target Geolocation Approximation (Optional)
Objective: Approximating geolocations (latitude and longitude) of specific targets using only imagery and telemetry data
- Apply one or more methods to all of the provided test image sequences (if possible, for comprehensive evaluation), or at least two sequences.
- Report results in a standardised format.
- Include a brief description of the methodology used, with results submitted in XML format.
Please click on the Author Instructions to see the detailed description of the challenges set in PETS 2025 and submission guidelines.
Please click on the Datasets link to see the detailed description of the datasets and how to download them.