Fisheye cameras in camera calibration block

[Erol444]

What does this PR do?

Before (current cam model)

After (with fisheye model enabled):

Type of Change

• New feature (non-breaking change that adds functionality)

Testing

• I have tested this change locally
• I have added/updated tests for this change

Test details:

Checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code where necessary, particularly in hard-to-understand areas
• My changes generate no new warnings or errors
• I have updated the documentation accordingly (if applicable)

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 91b618015b

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Keep fisheye remap arrays in float32

In the fisheye branch, x and y are computed from float32 grids using Python float intrinsics (cx, cy, fx, fy), which promotes them to float64; all downstream arrays (r2, discriminant, scale, map_x, map_y) then stay float64 as well. For high-resolution inputs (for example 4K), this creates several full-frame 8-byte temporaries and can spike per-frame memory into hundreds of MB, causing severe throughput drops or OOMs when use_fisheye_model is enabled. Casting intrinsics once to np.float32 (or otherwise forcing float32 math) avoids this regression.

Useful? React with 👍 / 👎.