TartanRGBT: Synchronized RGB, Thermal, and Depth Data for Robot Perception

TartanRGBT is a large-scale dataset containing synchronized RGB, thermal, and depth data collected across diverse indoor, outdoor, urban, off-road, and park environments. It is designed to support research in robot perception, visual localization, cross-modal representation learning, thermal vision, and multi-sensor fusion. The dataset is organized by day, trajectory, and modality and is authored by theairlabcmu.

Use Cases

• Training visual localization models based on synchronized RGB and depth data.
• Developing cross-modal representation learning algorithms using aligned RGB and thermal imagery.
• Researching thermal vision applications for robotics in diverse environments.
• Benchmarking multi-sensor fusion techniques for autonomous navigation.
• Evaluating robot perception systems in varied indoor and outdoor settings.

Strengths

• Large-scale dataset covering diverse environments including indoor, outdoor, urban, off-road, and park settings.
• Data is synchronized across RGB, thermal, and depth modalities, which is crucial for multi-sensor research.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Freshness should be verified as the last update timestamp is in the future (2026-05-16).

Provenance

Source

theairlabcmu

Collection Method

Collected across diverse environments; specific collection method is not detailed.

Freshness

Last updated 2026-05-16 19:53:26

Geography

Covers diverse indoor, outdoor, urban, off-road, and park environments; specific locations are not provided.