Add ChArUco board generator for dual-camera calibration jig#33
Open
danielkaijzer wants to merge 6 commits into
Open
Add ChArUco board generator for dual-camera calibration jig#33danielkaijzer wants to merge 6 commits into
danielkaijzer wants to merge 6 commits into
Conversation
Generates small/medium/large ChArUco boards (15/22/30 mm squares) as exact-size US-Letter PDFs for testing which marker size the headset cameras resolve at the jig working distance before committing to a laminated final board. Each board uses a distinct ID range so sizes don't collide. PDFs include a 50 mm reference ruler to catch print scaling. https://claude.ai/code/session_01LwRnc5nA2QKS7GYMUA8aHr
Each board now tiles as many squares as fit on US-Letter so the pattern covers the whole sheet. Camera sees more calibration points in every partial view. Grids are now 13x17 / 9x11 / 6x8 vs the previous 5x7. https://claude.ai/code/session_01LwRnc5nA2QKS7GYMUA8aHr
Cameras are more likely to see the lower part of the panel, so the board now fills from the bottom up and the 50 mm ruler + label sit in the top margin. https://claude.ai/code/session_01LwRnc5nA2QKS7GYMUA8aHr
Computes the rigid transform between the two non-overlapping cameras using the jig's two ChArUco boards. Formulated as a robot-world/hand-eye problem (cv2.calibrateRobotWorldHandEye) so it recovers both the camera extrinsic and the board-to-board transform from varied rig poses — no need to measure where the boards sit on the panels, only that the jig is rigid. Captures synchronized pairs from both cameras (SPACE/C/R/Q like the intrinsics script), solves each board pose via matchImagePoints+solvePnP, runs SHAH and LI and keeps the lower-residual result. Reports pose diversity, camera baseline, inter-camera angle, and recovered board separation (sanity-checked against measured panel spacing). Saves both T_eye_scene and T_scene_eye (the latter is what gaze projection needs). https://claude.ai/code/session_01LwRnc5nA2QKS7GYMUA8aHr
…flow Mirrors calibrate_scene_intrinsics.py for the eye camera (eye-cam resolution, optional vertical flip for inverted mounts) and saves eye_intrinsics.npz in the format the extrinsics script expects. Documents the full intrinsics + extrinsics calibration flow and output files in the backend README. https://claude.ai/code/session_01LwRnc5nA2QKS7GYMUA8aHr
Records the rig display's physical active area (~326.6 x 211.4 mm) and the logical-points-vs-native-pixels caveat, so the future data-collection pipeline can convert on-screen marker positions to metric object points and recover fixation depth via solvePnP. Reference data only; no pipeline code. https://claude.ai/code/session_01LwRnc5nA2QKS7GYMUA8aHr
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Add a script to generate printable ChArUco calibration boards at three different square sizes (15mm, 22mm, 30mm) for the dual-camera calibration jig. The boards are optimized to fill US Letter pages and include reference rulers for verification.
Key Changes
generate_charuco_boards.py: New script that:
Generated artifacts: Three pairs of PNG and PDF files (small, medium, large) ready for printing and use in calibration
Notable Implementation Details
reportlabfor PDF generation with precise millimeter-based positioninghttps://claude.ai/code/session_01LwRnc5nA2QKS7GYMUA8aHr