TartanRGBT Dataset

Overview

TartanRGBT is a large-scale, synchronized RGB–Thermal–Depth dataset collected across diverse indoor, outdoor, urban, off-road, and park environments. The dataset is designed to support research in robot perception, visual localization, cross-modal representation learning, thermal vision, and multi-sensor fusion.

All data is organized by day → trajectory (timestamped) → modality, distributed as ZIP archives to enable efficient storage and selective download.

If you use this dataset, please cite our work, AnyThermal: Towards Learning Universal Representations for Thermal Perception, accepted at ICRA 2026. Project website: https://anythermal.github.io/

@misc{maheshwari2026anythermallearninguniversalrepresentations,
      title={AnyThermal: Towards Learning Universal Representations for Thermal Perception},
      author={Parv Maheshwari and Jay Karhade and Yogesh Chawla and Isaiah Adu and Florian Heisen and Andrew Porco and Andrew Jong and Yifei Liu and Santosh Pitla and Sebastian Scherer and Wenshan Wang},
      year={2026},
      eprint={2602.06203},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2602.06203},
}

What's new in this release

This release corrects the earlier version by adding the per-frame flat-field-correction (FFC) flags, interpolated odometry poses, target timestamps, and TF static transforms that the AnyThermal dataset class expects. Image data is unchanged.

Trajectory folder names now follow the original capture-time naming (undistorted_images_all_cameras_<timestamp>) which matches the layout in custom_datasets/tartanRGBT/splits/sequence.yaml. The RGB modality folder is named rgb_in_thermal/ (not RGB_aligned_with_thermal/).

All FFC entries, target timestamps, and image frames are aligned at 1 Hz.

Sensors & Modalities

Each trajectory contains the following modality ZIP files:

Zip	Contents	Notes
rgb_in_thermal.zip	RGB images aligned to thermal frame	8-bit RGB
thermal_left_rect_8.zip	Left thermal, rectified, 8-bit	grayscale
thermal_right_rect_8.zip	Right thermal, rectified, 8-bit	grayscale
thermal_right_rect_16.zip	Right thermal, rectified, 16-bit	high dynamic range
zed_left_rect.zip	Left ZED RGB, rectified	from stereo rig
zed_right_rect.zip	Right ZED RGB, rectified	from stereo rig
stereo_depth.zip	Stereo depth maps	from ZED
metadata.zip	FFC flags, odometry, target_timestamps, tf, frame metadata	small (~5–10 MB / sequence)

All sequences are temporally aligned across modalities and sampled at 1 Hz.

Inside metadata.zip

Path	Description
thermal_left_ffc/data.txt	1 per kept frame: 0 = good frame, 1 = drop (during FFC recalibration)
thermal_right_ffc/data.txt	same, right cam
thermal_*_ffc/{errors,timestamps,frames}.{txt,yaml}	reference metadata (original capture-time files)
thermal_left_rect_16/{errors,timestamps,frames}.{txt,yaml}	reference metadata (16-bit left thermal not yet shipped here)
odometry/poses.npy	(N, 8) float array: [timestamp_ns, x, y, z, qx, qy, qz, qw]
tf/.../static_transform.txt	static inter-sensor transforms
target_timestamps.txt	image-aligned timestamps used by the AnyThermal odometry interpolation

Dataset Structure

After download and extraction, the dataset follows the structure below:

TartanRGBT_dataset/
├── day1/
│   └── undistorted_images_all_cameras_<timestamp>/
│       ├── rgb_in_thermal/                          # extracted from rgb_in_thermal.zip
│       │   ├── 00000000_rgb_in_thermal.png
│       │   ├── 00000010_rgb_in_thermal.png
│       │   └── ...
│       ├── thermal_left_rect_8/
│       ├── thermal_right_rect_8/
│       ├── thermal_right_rect_16/
│       ├── zed_left_rect/
│       ├── zed_right_rect/
│       ├── stereo_depth/
│       ├── thermal_left_ffc/data.txt                # extracted from metadata.zip
│       ├── thermal_right_ffc/data.txt
│       ├── odometry/poses.npy
│       ├── tf/...
│       ├── target_timestamps.txt
│       └── [*.zip]                                  # optional: see --delete_zips flag
├── day2/, day3/, day4/, day5/
├── calibration/
├── AnyThermal_data_distribution.csv
├── data_extraction.py
└── README.md

On HuggingFace, only ZIPs + small root files are stored, preserving the same hierarchy: dayX/<trajectory>/*.zip.

Requirements

pip install huggingface_hub

Extraction

The data_extraction.py script downloads every ZIP from this repository and extracts each ZIP in place. After extraction, the layout above is produced.

python data_extraction.py <base_dir> [--delete_zips yes|no]

Arguments

• base_dir — directory where TartanRGBT_dataset/ will be created.
• --delete_zips (default: no) — whether to delete the source ZIPs after extraction succeeds.

Quick sanity check post-extraction

import os, numpy as np
ROOT = "<base_dir>/TartanRGBT_dataset"
seq = "day1/undistorted_images_all_cameras_20250822_115703"
print("rgb frames:    ", len(os.listdir(f"{ROOT}/{seq}/rgb_in_thermal/")))
print("thermal frames:", len(os.listdir(f"{ROOT}/{seq}/thermal_left_rect_8/")))
with open(f"{ROOT}/{seq}/thermal_left_ffc/data.txt") as f:
    ffc = [int(x.strip()) for x in f]
print("ffc length:    ", len(ffc), "  n_dropped:", sum(ffc))
print("ts lines:      ", sum(1 for _ in open(f"{ROOT}/{seq}/target_timestamps.txt")))
poses = np.load(f"{ROOT}/{seq}/odometry/poses.npy")
print("poses shape:   ", poses.shape)

You should see rgb frames == thermal frames == ffc length == ts lines (all 1 Hz aligned).