test: use cow array in iou

bindings/src/lib.rs

@@ -2,6 +2,7 @@ mod utils;
 
 use std::fmt::Debug;
 
+use ndarray::Array1;
 use num_traits::{Bounded, Float, Num, Signed, ToPrimitive};
 use numpy::{PyArray1, PyArray2, PyArray3};
 use powerboxesrs::{boxes, diou, giou, iou, nms, tiou};
@@ -123,7 +124,7 @@ fn _powerboxes(_py: Python, m: &PyModule) -> PyResult<()> {
 #[pyfunction]
 fn masks_to_boxes(_py: Python, masks: &PyArray3<bool>) -> PyResult<Py<PyArray2<usize>>> {
     let masks = preprocess_array3(masks);
-    let boxes = boxes::masks_to_boxes(&masks);
+    let boxes = boxes::masks_to_boxes(masks);
     let boxes_as_numpy = utils::array_to_numpy(_py, boxes).unwrap();
     return Ok(boxes_as_numpy.to_owned());
 }
@@ -179,7 +180,7 @@ fn diou_distance_generic<T>(
     boxes2: &PyArray2<T>,
 ) -> PyResult<Py<PyArray2<f64>>>
 where
-    T: Float + numpy::Element,
+    T: Num + Float + numpy::Element,
 {
     let boxes1 = preprocess_boxes(boxes1).unwrap();
     let boxes2 = preprocess_boxes(boxes2).unwrap();
@@ -216,7 +217,7 @@ where
 {
     let boxes1 = preprocess_boxes(boxes1).unwrap();
     let boxes2 = preprocess_boxes(boxes2).unwrap();
-    let iou = iou::iou_distance(&boxes1, &boxes2);
+    let iou = iou::iou_distance(boxes1.to_owned(), boxes2.to_owned());
     let iou_as_numpy = utils::array_to_numpy(_py, iou).unwrap();
     return Ok(iou_as_numpy.to_owned());
 }
@@ -806,7 +807,7 @@ where
             ))
         }
     };
-    let converted_boxes = boxes::box_convert(&boxes, &in_fmt, &out_fmt);
+    let converted_boxes = boxes::box_convert(&boxes, in_fmt, out_fmt);
     let converted_boxes_as_numpy = utils::array_to_numpy(_py, converted_boxes).unwrap();
     return Ok(converted_boxes_as_numpy.to_owned());
 }
@@ -902,12 +903,13 @@ fn nms_generic<T>(
     score_threshold: f64,
 ) -> PyResult<Py<PyArray1<usize>>>
 where
-    T: Num + numpy::Element + PartialOrd + ToPrimitive + Copy,
+    T: numpy::Element + Num + PartialEq + PartialOrd + ToPrimitive + Copy,
 {
     let boxes = preprocess_boxes(boxes).unwrap();
     let scores = preprocess_array1(scores);
     let keep = nms::nms(&boxes, &scores, iou_threshold, score_threshold);
-    let keep_as_numpy = utils::array_to_numpy(_py, keep).unwrap();
+    let keep_as_ndarray = Array1::from(keep);
+    let keep_as_numpy = utils::array_to_numpy(_py, keep_as_ndarray).unwrap();
     return Ok(keep_as_numpy.to_owned());
 }
 #[pyfunction]
@@ -1064,21 +1066,23 @@ fn rtree_nms_generic<T>(
     score_threshold: f64,
 ) -> PyResult<Py<PyArray1<usize>>>
 where
-    T: Num
-        + numpy::Element
-        + PartialOrd
-        + ToPrimitive
-        + Copy
+    T: numpy::Element
+        + Num
         + Signed
         + Bounded
         + Debug
+        + PartialEq
+        + PartialOrd
+        + ToPrimitive
+        + Copy
         + Sync
         + Send,
 {
     let boxes = preprocess_boxes(boxes).unwrap();
     let scores = preprocess_array1(scores);
     let keep = nms::rtree_nms(&boxes, &scores, iou_threshold, score_threshold);
-    let keep_as_numpy = utils::array_to_numpy(_py, keep).unwrap();
+    let keep_as_ndarray = Array1::from(keep);
+    let keep_as_numpy = utils::array_to_numpy(_py, keep_as_ndarray).unwrap();
     return Ok(keep_as_numpy.to_owned());
 }
 #[pyfunction]

powerboxesrs/src/iou.rs

@@ -3,7 +3,7 @@ use crate::{
     rotation::{intersection_area, minimal_bounding_rect, Rect},
     utils,
 };
-use ndarray::{Array2, Zip};
+use ndarray::{Array2, ArrayView2, CowArray, Dim, Zip};
 use num_traits::{Num, ToPrimitive};
 use rstar::RTree;
 
@@ -29,10 +29,13 @@ use rstar::RTree;
 /// let iou = iou_distance(&boxes1, &boxes2);
 /// assert_eq!(iou, array![[0.8571428571428572, 1.],[1., 0.8571428571428572]]);
 /// ```
-pub fn iou_distance<N>(boxes1: &Array2<N>, boxes2: &Array2<N>) -> Array2<f64>
+pub fn iou_distance<'a, N, BA>(boxes1: BA, boxes2: BA) -> Array2<f64>
 where
-    N: Num + PartialOrd + ToPrimitive + Copy,
+    N: Num + PartialEq + PartialOrd + ToPrimitive + Copy + 'a,
+    BA: Into<CowArray<'a, N, Dim<[usize; 2]>>>,
 {
+    let boxes1 = boxes1.into();
+    let boxes2 = boxes2.into();
     let num_boxes1 = boxes1.nrows();
     let num_boxes2 = boxes2.nrows();
 
@@ -95,10 +98,13 @@ where
 /// let iou = parallel_iou_distance(&boxes1, &boxes2);
 /// assert_eq!(iou, array![[0.8571428571428572, 1.],[1., 0.8571428571428572]]);
 /// ```
-pub fn parallel_iou_distance<N>(boxes1: &Array2<N>, boxes2: &Array2<N>) -> Array2<f64>
+pub fn parallel_iou_distance<'a, N, BA>(boxes1: BA, boxes2: BA) -> Array2<f64>
 where
-    N: Num + PartialOrd + ToPrimitive + Copy + Clone + Sync + Send,
+    N: Num + PartialEq + PartialOrd + ToPrimitive + Send + Sync + Copy + 'a,
+    BA: Into<ArrayView2<'a, N>>,
 {
+    let boxes1 = boxes1.into();
+    let boxes2 = boxes2.into();
     let num_boxes1 = boxes1.nrows();
     let num_boxes2 = boxes2.nrows();
 
@@ -151,7 +157,13 @@ where
 /// # Returns
 /// A 2D array containing the Rotated IoU distance matrix. The element at position (i, j) represents
 /// the Rotated IoU distance between the i-th box in `boxes1` and the j-th box in `boxes2`.
-pub fn rotated_iou_distance(boxes1: &Array2<f64>, boxes2: &Array2<f64>) -> Array2<f64> {
+pub fn rotated_iou_distance<'a, BA>(boxes1: BA, boxes2: BA) -> Array2<f64>
+where
+    BA: Into<ArrayView2<'a, f64>>,
+{
+    let boxes1 = boxes1.into();
+    let boxes2 = boxes2.into();
+
     let num_boxes1 = boxes1.nrows();
     let num_boxes2 = boxes2.nrows();