[aswanth-07]

Description

Detections.box_aspect_ratio can return a negative ratio for an ordered integer box because it subtracts corners in the coordinate dtype. For example:

import numpy as np
import supervision as sv

detections = sv.Detections(xyxy=np.array([[-120, 0, 120, 20]], dtype=np.int8))
print(detections.box_aspect_ratio)  # [-0.8], expected [12.0]

Motivation and Context

I found this while checking aspect-ratio filtering on integer detections. Overflow in either width or height can make the filter select the wrong boxes. box_area already handles the same arithmetic boundary.

Changes Made

• Widen integer coordinates smaller than 64 bits to int64 and retain Python-integer arithmetic for int64/uint64, then convert the differences to float64. This preserves narrow boxes whose origins exceed float64's exact-integer range.
• Add regression and compatibility coverage plus an Unreleased changelog entry. Floating arithmetic, float64 ratios, zero-height NaNs and input coordinates stay unchanged.

Testing

• pytest tests/detection/test_core.py -q: 263 passed.

• pytest -q --cov=supervision: 5876 passed, six skipped, with 94% coverage after merging current develop. The four signed-range regressions fail against the pristine property; the other 12 focused cases pass.

• pre-commit run --all-files: all hooks passed.

• Additional validation: 99 exact pristine/patch compatibility checks, 16 integer-range checks against Python Fraction arithmetic, and 24 previous-head parity checks for extreme, strided and read-only integer arrays.

• Added/updated tests, and the full suite passes locally

• Updated docs (docstrings / mkdocs entry) for new or changed public API

• Added a changelog entry in docs/changelog.md under Unreleased (skip for lint/type/format-only or pure doc changes)

Additional Notes

The public API and stored arrays do not change. Dtype-based widening follows get_bbox_size_category; converting corners directly to float64 would lose one-unit widths above 2**53.

AI assistance disclosure: AI-assisted development tools were used during investigation and implementation. The reported tests were run against the final diff.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 94%. Comparing base (13ef387) to head (afcdef8).

Additional details and impacted files

@@           Coverage Diff           @@
##           develop   #2729   +/-   ##
=======================================
- Coverage       94%     94%   -0%     
=======================================
  Files           83      83           
  Lines        13316   13322    +6     
=======================================
+ Hits         12509   12512    +3     
- Misses         807     810    +3     

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Copilot]

🟡 Changes recommended

Integer arrays below 64-bit unnecessarily use costly object arithmetic, and the changelog metadata remains stale.

2 open findings

• Avoid object conversion for narrow integer bounding-box arrays · New
• Update stale date_modified in changelog front matter · New

What changed in this PR

Fixes integer overflow in Detections.box_aspect_ratio.

Changes:

• Uses exact integer subtraction before float conversion.
• Adds boundary and compatibility tests.
• Documents the fix in the changelog.

File	Description
src/​supervision/​detection/​core.py	Corrects integer aspect-ratio arithmetic.
tests/​detection/​test_core.py	Adds regression and edge-case coverage.
docs/​changelog.md	Records the behavior fix.

🧠 Review effort: Balanced

---

💡 Configure MCP servers for context-aware, tailored reviews. Learn more in the docs.