[Add] Add rotate iou cpu evaluation on kitti dateset

Author: shun001

mmdet3d/evaluation/functional/kitti_utils/eval.py

@@ -4,6 +4,7 @@
 
 import numba
 import numpy as np
+import torch
 
 
 @numba.jit
@@ -115,8 +116,12 @@ def image_box_overlap(boxes, query_boxes, criterion=-1):
 
 
 def bev_box_overlap(boxes, qboxes, criterion=-1):
-    from .rotate_iou import rotate_iou_gpu_eval
-    riou = rotate_iou_gpu_eval(boxes, qboxes, criterion)
+    if torch.cuda.is_available():
+        from .rotate_iou import rotate_iou_gpu_eval
+        riou = rotate_iou_gpu_eval(boxes, qboxes, criterion)
+    else:
+        from .rotate_iou_cpu import rotate_iou_cpu_eval
+        riou = rotate_iou_cpu_eval(boxes, qboxes, criterion)
     return riou
 
 
@@ -153,9 +158,14 @@ def d3_box_overlap_kernel(boxes, qboxes, rinc, criterion=-1):
 
 
 def d3_box_overlap(boxes, qboxes, criterion=-1):
-    from .rotate_iou import rotate_iou_gpu_eval
-    rinc = rotate_iou_gpu_eval(boxes[:, [0, 2, 3, 5, 6]],
-                               qboxes[:, [0, 2, 3, 5, 6]], 2)
+    if torch.cuda.is_available():
+        from .rotate_iou import rotate_iou_gpu_eval
+        rinc = rotate_iou_gpu_eval(boxes[:, [0, 2, 3, 5, 6]],
+                                   qboxes[:, [0, 2, 3, 5, 6]], 2)
+    else:
+        from .rotate_iou_cpu import rotate_iou_cpu_eval
+        rinc = rotate_iou_cpu_eval(boxes[:, [0, 2, 3, 5, 6]],
+                                   qboxes[:, [0, 2, 3, 5, 6]], 2)
     d3_box_overlap_kernel(boxes, qboxes, rinc, criterion)
     return rinc
 

mmdet3d/evaluation/functional/kitti_utils/rotate_iou_cpu.py

@@ -0,0 +1,254 @@
+# Copyright (c) OpenMMLab. All rights reserved.
+#####################
+# Based on https://github.com/hongzhenwang/RRPN-revise
+# Licensed under The MIT License
+# Author: yanyan, [email protected]
+#####################
+import math
+
+import numpy as np
+
+
+def div_up(m, n):
+    return m // n + (m % n > 0)
+
+
+def trangle_area(a, b, c):
+    return ((a[0] - c[0]) * (b[1] - c[1]) - (a[1] - c[1]) *
+            (b[0] - c[0])) / 2.0
+
+
+def area(int_pts, num_of_inter):
+    area_val = 0.0
+    for i in range(num_of_inter - 2):
+        area_val += abs(
+            trangle_area(int_pts[:2], int_pts[2 * i + 2:2 * i + 4],
+                         int_pts[2 * i + 4:2 * i + 6]))
+    return area_val
+
+
+def sort_vertex_in_convex_polygon(int_pts, num_of_inter):
+    if num_of_inter > 0:
+        center = np.zeros((2, ), dtype=np.float32)
+        center[:] = 0.0
+        for i in range(num_of_inter):
+            center[0] += int_pts[2 * i]
+            center[1] += int_pts[2 * i + 1]
+        center[0] /= num_of_inter
+        center[1] /= num_of_inter
+        v = np.zeros((2, ), dtype=np.float32)
+        vs = np.zeros((16, ), dtype=np.float32)
+        for i in range(num_of_inter):
+            v[0] = int_pts[2 * i] - center[0]
+            v[1] = int_pts[2 * i + 1] - center[1]
+            d = math.sqrt(v[0] * v[0] + v[1] * v[1])
+            v[0] = v[0] / d
+            v[1] = v[1] / d
+            if v[1] < 0:
+                v[0] = -2 - v[0]
+            vs[i] = v[0]
+        j = 0
+        temp = 0
+        for i in range(1, num_of_inter):
+            if vs[i - 1] > vs[i]:
+                temp = vs[i]
+                tx = int_pts[2 * i]
+                ty = int_pts[2 * i + 1]
+                j = i
+                while j > 0 and vs[j - 1] > temp:
+                    vs[j] = vs[j - 1]
+                    int_pts[j * 2] = int_pts[j * 2 - 2]
+                    int_pts[j * 2 + 1] = int_pts[j * 2 - 1]
+                    j -= 1
+
+                vs[j] = temp
+                int_pts[j * 2] = tx
+                int_pts[j * 2 + 1] = ty
+
+
+def line_segment_intersection(pts1, pts2, i, j, temp_pts):
+    A = np.zeros((2, ), dtype=np.float32)
+    B = np.zeros((2, ), dtype=np.float32)
+    C = np.zeros((2, ), dtype=np.float32)
+    D = np.zeros((2, ), dtype=np.float32)
+
+    A[0] = pts1[2 * i]
+    A[1] = pts1[2 * i + 1]
+
+    B[0] = pts1[2 * ((i + 1) % 4)]
+    B[1] = pts1[2 * ((i + 1) % 4) + 1]
+
+    C[0] = pts2[2 * j]
+    C[1] = pts2[2 * j + 1]
+
+    D[0] = pts2[2 * ((j + 1) % 4)]
+    D[1] = pts2[2 * ((j + 1) % 4) + 1]
+    BA0 = B[0] - A[0]
+    BA1 = B[1] - A[1]
+    DA0 = D[0] - A[0]
+    CA0 = C[0] - A[0]
+    DA1 = D[1] - A[1]
+    CA1 = C[1] - A[1]
+    acd = DA1 * CA0 > CA1 * DA0
+    bcd = (D[1] - B[1]) * (C[0] - B[0]) > (C[1] - B[1]) * (D[0] - B[0])
+    if acd != bcd:
+        abc = CA1 * BA0 > BA1 * CA0
+        abd = DA1 * BA0 > BA1 * DA0
+        if abc != abd:
+            DC0 = D[0] - C[0]
+            DC1 = D[1] - C[1]
+            ABBA = A[0] * B[1] - B[0] * A[1]
+            CDDC = C[0] * D[1] - D[0] * C[1]
+            DH = BA1 * DC0 - BA0 * DC1
+            Dx = ABBA * DC0 - BA0 * CDDC
+            Dy = ABBA * DC1 - BA1 * CDDC
+            temp_pts[0] = Dx / DH
+            temp_pts[1] = Dy / DH
+            return True
+    return False
+
+
+def line_segment_intersection_v1(pts1, pts2, i, j, temp_pts):
+    a = np.zeros((2, ), dtype=np.float32)
+    b = np.zeros((2, ), dtype=np.float32)
+    c = np.zeros((2, ), dtype=np.float32)
+    d = np.zeros((2, ), dtype=np.float32)
+
+    a[0] = pts1[2 * i]
+    a[1] = pts1[2 * i + 1]
+
+    b[0] = pts1[2 * ((i + 1) % 4)]
+    b[1] = pts1[2 * ((i + 1) % 4) + 1]
+
+    c[0] = pts2[2 * j]
+    c[1] = pts2[2 * j + 1]
+
+    d[0] = pts2[2 * ((j + 1) % 4)]
+    d[1] = pts2[2 * ((j + 1) % 4) + 1]
+
+    area_abc = trangle_area(a, b, c)
+    area_abd = trangle_area(a, b, d)
+
+    if area_abc * area_abd >= 0:
+        return False
+
+    area_cda = trangle_area(c, d, a)
+    area_cdb = area_cda + area_abc - area_abd
+
+    if area_cda * area_cdb >= 0:
+        return False
+    t = area_cda / (area_abd - area_abc)
+
+    dx = t * (b[0] - a[0])
+    dy = t * (b[1] - a[1])
+    temp_pts[0] = a[0] + dx
+    temp_pts[1] = a[1] + dy
+    return True
+
+
+def point_in_quadrilateral(pt_x, pt_y, corners):
+    ab0 = corners[2] - corners[0]
+    ab1 = corners[3] - corners[1]
+
+    ad0 = corners[6] - corners[0]
+    ad1 = corners[7] - corners[1]
+
+    ap0 = pt_x - corners[0]
+    ap1 = pt_y - corners[1]
+
+    abab = ab0 * ab0 + ab1 * ab1
+    abap = ab0 * ap0 + ab1 * ap1
+    adad = ad0 * ad0 + ad1 * ad1
+    adap = ad0 * ap0 + ad1 * ap1
+
+    return abab >= abap and abap >= 0 and adad >= adap and adap >= 0
+
+
+def quadrilateral_intersection(pts1, pts2, int_pts):
+    num_of_inter = 0
+    for i in range(4):
+        if point_in_quadrilateral(pts1[2 * i], pts1[2 * i + 1], pts2):
+            int_pts[num_of_inter * 2] = pts1[2 * i]
+            int_pts[num_of_inter * 2 + 1] = pts1[2 * i + 1]
+            num_of_inter += 1
+        if point_in_quadrilateral(pts2[2 * i], pts2[2 * i + 1], pts1):
+            int_pts[num_of_inter * 2] = pts2[2 * i]
+            int_pts[num_of_inter * 2 + 1] = pts2[2 * i + 1]
+            num_of_inter += 1
+    temp_pts = np.zeros((2, ), dtype=np.float32)
+    for i in range(4):
+        for j in range(4):
+            has_pts = line_segment_intersection(pts1, pts2, i, j, temp_pts)
+            if has_pts:
+                int_pts[num_of_inter * 2] = temp_pts[0]
+                int_pts[num_of_inter * 2 + 1] = temp_pts[1]
+                num_of_inter += 1
+
+    return num_of_inter
+
+
+def rbbox_to_corners(corners, rbbox):
+    # generate clockwise corners and rotate it clockwise
+    angle = rbbox[4]
+    a_cos = math.cos(angle)
+    a_sin = math.sin(angle)
+    center_x = rbbox[0]
+    center_y = rbbox[1]
+    x_d = rbbox[2]
+    y_d = rbbox[3]
+    corners_x = np.zeros((4, ), dtype=np.float32)
+    corners_y = np.zeros((4, ), dtype=np.float32)
+    corners_x[0] = -x_d / 2
+    corners_x[1] = -x_d / 2
+    corners_x[2] = x_d / 2
+    corners_x[3] = x_d / 2
+    corners_y[0] = -y_d / 2
+    corners_y[1] = y_d / 2
+    corners_y[2] = y_d / 2
+    corners_y[3] = -y_d / 2
+    for i in range(4):
+        corners[2 * i] = a_cos * corners_x[i] + a_sin * corners_y[i] + center_x
+        corners[2 * i +
+                1] = -a_sin * corners_x[i] + a_cos * corners_y[i] + center_y
+
+
+def inter(rbbox1, rbbox2):
+    corners1 = np.zeros((8, ), dtype=np.float32)
+    corners2 = np.zeros((8, ), dtype=np.float32)
+    intersection_corners = np.zeros((16, ), dtype=np.float32)
+
+    rbbox_to_corners(corners1, rbbox1)
+    rbbox_to_corners(corners2, rbbox2)
+
+    num_intersection = quadrilateral_intersection(corners1, corners2,
+                                                  intersection_corners)
+    sort_vertex_in_convex_polygon(intersection_corners, num_intersection)
+    # print(intersection_corners.reshape([-1, 2])[:num_intersection])
+
+    return area(intersection_corners, num_intersection)
+
+
+def devRotateIoUEval(rbox1, rbox2, criterion=-1):
+    area1 = rbox1[2] * rbox1[3]
+    area2 = rbox2[2] * rbox2[3]
+    area_inter = inter(rbox1, rbox2)
+    if criterion == -1:
+        return area_inter / (area1 + area2 - area_inter)
+    elif criterion == 0:
+        return area_inter / area1
+    elif criterion == 1:
+        return area_inter / area2
+    else:
+        return area_inter
+
+
+def rotate_iou_cpu_eval(dev_boxes, dev_query_boxes, criterion=-1):
+    num_boxes = dev_boxes.shape[0]
+    num_qboxes = dev_query_boxes.shape[0]
+    dev_iou = np.zeros((num_boxes, num_qboxes))
+    for box_i in range(num_boxes):
+        for qbox_i in range(num_qboxes):
+            dev_iou[box_i,
+                    qbox_i] = devRotateIoUEval(dev_query_boxes[qbox_i],
+                                               dev_boxes[box_i], criterion)
+    return dev_iou

tests/test_evaluation/test_functional/test_kitti_eval.py

@@ -1,14 +1,10 @@
 # Copyright (c) OpenMMLab. All rights reserved.
 import numpy as np
-import pytest
-import torch
 
 from mmdet3d.evaluation import do_eval, eval_class, kitti_eval
 
 
 def test_do_eval():
-    if not torch.cuda.is_available():
-        pytest.skip('test requires GPU and CUDA')
     gt_name = np.array(
         ['Pedestrian', 'Cyclist', 'Car', 'Car', 'Car', 'DontCare', 'DontCare'])
     gt_truncated = np.array([0., 0., 0., -1., -1., -1., -1.])
@@ -128,8 +124,6 @@ def test_do_eval():
 
 
 def test_kitti_eval():
-    if not torch.cuda.is_available():
-        pytest.skip('test requires GPU and CUDA')
     gt_name = np.array(
         ['Pedestrian', 'Cyclist', 'Car', 'Car', 'Car', 'DontCare', 'DontCare'])
     gt_truncated = np.array([0., 0., 0., -1., -1., -1., -1.])