Timed Camera-LiDAR ordered-batch comparison

This archive contains all six fresh-process runs, complete prepared problems, source/helper/schema pins, results and receipts for the linked report. index.json binds every publication byte other than itself.

The nine scans and 9,216 original points of KITTI 2011_09_30_drive_0018_sync are unchanged from the previous evaluation-reuse control. Original point distributions and OXTS velocities are preserved in source-points.json.gz and motion-control.json.gz. input-provenance.json and control-generator.py.txt declare their lineage, point times, image means, camera/extrinsics and fixed noise. Generated image matches and point times make this a performance/parity control, not independent accuracy evidence.

KITTI attribution: Andreas Geiger, Philip Lenz, Christoph Stiller and Raquel Urtasun. Public KITTI data and its derived control retain CC BY-NC-SA 3.0: https://www.cvlibs.net/datasets/kitti/ and https://creativecommons.org/licenses/by-nc-sa/3.0/. Calibrex source snapshots and measurement code retain Apache-2.0.

The protocol uses benchmark schema v0.2. Observed scalar/batch work is checked against an independent replay of the unchanged trace and two-clock cache. All 556,032 deskewed points and 2,462,720 projections remain in each run. Only top-level result provenance created_at/git_commit are excluded from numerical comparisons; actual full/short Git IDs and source SHA are checked separately. Prior v0.1 archives remain independently auditable.

Run python -m tools.benchmark_continuous_camera_lidar_reuse --audit-only docs/assets/continuous_camera_lidar_batching from the repository root. Frozen worker paths in the historical protocol describe the original run; the audit uses the portable capsules and needs no original path or download.