TartanRGBT: RGB-Depth-Thermal Simulation Dataset for Robotics

TartanRGBT is a FiftyOne dataset containing 5,952 samples. The dataset was created by harpreetsahota and was last updated on HuggingFace in May 2026. The full description is available on the dataset's page.

Use Cases

• Train sensor fusion models based on the multimodal RGB, depth, and thermal data.
• Benchmark visual perception algorithms for autonomous navigation in simulated environments.
• Develop and test robotic vision systems using synthetic data before real-world deployment.

Strengths

• Contains 5,952 samples, providing a substantial collection for training and evaluation.
• The dataset is multimodal, likely containing RGB, depth, and thermal imagery as suggested by its name.

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).

Provenance

Source

harpreetsahota on HuggingFace

Freshness

Last updated 2026-05-23 03:32:36