Find a Planar Marker
A square marker beside an exhibit is visible in the guide camera.
Read the full lesson and open its transcript below. A narrated video will be added after production.
What you will learn
A square marker beside an exhibit is visible in the guide camera.
A practical question
A printed square marker is attached to a flat wall. In the camera image it appears skewed. If we know its real corner layout, detecting its four image corners gives us useful geometry.
The useful idea
Check that all corners belong to the same marker and that its identity and physical size are known. A false corner on a neighboring poster can produce a plausible but wrong mapping. Lens distortion and a partly hidden corner can also disturb the result.
Work through the example
Overlay the detected outline on the raw frame and compare the projected interior pattern. Repeat under shadow and at an angle. The marker is a controlled reference; its detection still has uncertainty. A planar marker has four visible corners. Detect them, confirm their order around the shape, and estimate the mapping only after rejecting an implausible arrangement. A swapped corner can send a projected graphic across the camera view even if the math routine returns a result. Overlay the mapped outline on the original marker and inspect it near all four edges. Save the detected corners so a teammate can reproduce a failed placement. Use this case to test your understanding: Detect four corners and check their ordering before estimating the mapping. Here is the limitation to watch for: A partially hidden or mistaken marker gives unstable geometry. The answer follows from that evidence: Known planar marker corners provide correspondences between marker coordinates and the image. Before placing an overlay, number the marker corners in order around its edge. Project them back after fitting the mapping. If one corner jumps across the marker, reject the pose instead of smoothing the graphic until it looks acceptable. The corner order must remain consistent from capture to projection.
Limits and next step
With reliable image points and known marker coordinates, we can estimate the camera's pose relative to it. That's our next step.
Worked example
Detect four corners and check their ordering before estimating the mapping.
Quick check: test your understanding
Try it yourself: Consider this case: Detect four corners and check their ordering before estimating the mapping. Explain how you would check the result. In your answer, address this limitation: A partially hidden or mistaken marker gives unstable geometry.
Check the reasoning
Known planar marker corners provide correspondences between marker coordinates and the image. A partially hidden or mistaken marker gives unstable geometry.
What is the main idea behind find a planar marker?
Known planar marker corners provide correspondences between marker coordinates and the image.
How is it used in the example?
Detect four corners and check their ordering before estimating the mapping.
What limitation should be checked?
A partially hidden or mistaken marker gives unstable geometry.
Further reading
Programming Computer Vision with Python — Jan Erik Solem.
Original ToolsInVision teaching examples. Editorial approach · Suggest a correction
Accessible referenceLesson 61 transcript303 words · Narration ready for the upcoming videoOpen transcript
A printed square marker is attached to a flat wall. In the camera image it appears skewed. If we know its real corner layout, detecting its four image corners gives us useful geometry.
Check that all corners belong to the same marker and that its identity and physical size are known. A false corner on a neighboring poster can produce a plausible but wrong mapping. Lens distortion and a partly hidden corner can also disturb the result.
Overlay the detected outline on the raw frame and compare the projected interior pattern. Repeat under shadow and at an angle. The marker is a controlled reference; its detection still has uncertainty. A planar marker has four visible corners. Detect them, confirm their order around the shape, and estimate the mapping only after rejecting an implausible arrangement. A swapped corner can send a projected graphic across the camera view even if the math routine returns a result. Overlay the mapped outline on the original marker and inspect it near all four edges. Save the detected corners so a teammate can reproduce a failed placement. Use this case to test your understanding: Detect four corners and check their ordering before estimating the mapping. Here is the limitation to watch for: A partially hidden or mistaken marker gives unstable geometry. The answer follows from that evidence: Known planar marker corners provide correspondences between marker coordinates and the image. Before placing an overlay, number the marker corners in order around its edge. Project them back after fitting the mapping. If one corner jumps across the marker, reject the pose instead of smoothing the graphic until it looks acceptable. The corner order must remain consistent from capture to projection.
With reliable image points and known marker coordinates, we can estimate the camera's pose relative to it. That's our next step.