[Zador-Pataki]

Motivation

Add more datasets and camera setups to the standard benchmark commands.

Changes

• LaMAR: use complete trajectories with available ground truth by default; keep sample-300 for the paper trajectories.
• Oxford Spires: rectified monocular, native fisheye and three-camera rig modes across 13 scenes.
• Hilti: 360° panoramas from 30 sequences.
• EuRoC / ETH3D-SLAM: stereo preparation that reuses existing cam0 images.

Includes source inventories, preparation, benchmark definitions and usage instructions. Hilti stitching can introduce near-field parallax.

Benchmark results: additional datasets

LaMAR results cover 53 campaign trajectories, versus 128 recordings in the new preparation. Spires monocular results cover 10 of 13 scenes.

LaMAR complete · 53 scenes · W-AUC at 5%

Method	ATE (m) ↓	10 m	25 m	50 m	100 m	Full
RoMaV2	1.452	92.42	92.83	92.38	92.59	89.76

Spires rectified monocular · 10 scenes · W-AUC at 5%

Method	ATE (m) ↓	10 m	25 m	50 m	100 m	Full
RoMaV2	0.130	98.98	99.10	99.11	99.10	99.54

Hilti forward monocular · 30 scenes · W-AUC at 5%

Method	ATE (m) ↓	10 m	25 m	50 m	100 m	Full
RoMaV2	0.822	93.96	93.96	93.29	92.44	94.46

Hilti three-view · 30 scenes · W-AUC at 5%

Method	ATE (m) ↓	10 m	25 m	50 m	100 m	Full
RoMaV2	0.206	94.80	96.22	96.94	97.21	98.13

Spires three-camera rig · 13 scenes · W-AUC at 5%

Method	ATE (m) ↓	10 m	25 m	50 m	100 m	Full
RoMaV2	0.119	99.15	99.24	99.26	99.28	99.59