Add comprehensive unit tests and benchmarks for both FAST and ORB.
- For FAST:
- test detection on synthetic/image patterns
- test NMS correctness
- For ORB:
- test pyramid construction, angle computation, and descriptor length
- property-based/edge-case tests
Add benches/features2d_bench.rs using Criterion.
fn bench_fast(c: &mut Criterion) {
// ...
}
fn bench_orb_detect_and_compute(c: &mut Criterion) {
// ...
}
Compare results to OpenCV where practical (but do not require exact parity).
References: https://docs.opencv.org/4.10.0/d5/d51/group__features2d__main.html
Add comprehensive unit tests and benchmarks for both FAST and ORB.
Add
benches/features2d_bench.rsusing Criterion.Compare results to OpenCV where practical (but do not require exact parity).
References: https://docs.opencv.org/4.10.0/d5/d51/group__features2d__main.html