Update chamfer_distance.py

hy3dshape/tools/evaluation/chamfer_distance.py

@@ -9,6 +9,30 @@ except Exception:
     except Exception:
         raise ImportError("Requires scipy.spatial.cKDTree or sklearn.neighbors. Install scipy or scikit-learn.")
 
+def normalize_point_cloud_dimension(points):
+    """
+    将点云数据按维度独立归一化到[-1, 1]范围。
+    
+    参数:
+        points (np.ndarray): 输入点云，形状为(n, 3)。
+        
+    返回:
+        np.ndarray: 归一化后的点云。
+        tuple: 每个维度的最小值，用于反归一化。
+        tuple: 每个维度的最大值，用于反归一化。
+    """
+    # 计算每个维度（列）的最小值和最大值
+    min_vals = np.min(points, axis=0)
+    max_vals = np.max(points, axis=0)
+    
+    # 计算每个维度的范围，避免除以零
+    ranges = max_vals - min_vals
+    ranges[ranges == 0] = 1e-8  # 如果某个维度值全相同，则范围设为一个小值
+    
+    normalized_points = (points - min_vals) / ranges  # 先归一化到[0,1]
+    normalized_points = normalized_points * 2 - 1    # 再映射到[-1,1]
+    
+    return normalized_points, min_vals, max_vals
 
 def sample_points_from_mesh(vertices: np.ndarray, faces: np.ndarray, n_samples: int) -> np.ndarray:
     """
@@ -21,6 +45,7 @@ def sample_points_from_mesh(vertices: np.ndarray, faces: np.ndarray, n_samples:
     Returns: (n_samples, 3) sampled points (float32)
     """
     v = np.asarray(vertices).reshape(-1, 3).astype(np.float64)
+    v, _, _ = normalize_point_cloud_dimension(v)
     f = np.asarray(faces).reshape(-1, 3).astype(np.int64)
 
     v0 = v[f[:, 0], :]
@@ -116,15 +141,8 @@ def chamfer_distance_from_meshes(pred_vertices: np.ndarray,
     B_to_A_l2 = float(np.mean(d_gt_to_pred))
     cd_l2 = 0.5 * (A_to_B_l2 + B_to_A_l2)
 
-    A_to_B_l2_sq = float(np.mean(d_pred_to_gt ** 2))
-    B_to_A_l2_sq = float(np.mean(d_gt_to_pred ** 2))
-    cd_l2_sq = 0.5 * (A_to_B_l2_sq + B_to_A_l2_sq)
-
     metrics = {
-        'cd_l2_sq': cd_l2_sq,
         'cd_l2': cd_l2,
-        'A_to_B_l2_sq': A_to_B_l2_sq,
-        'B_to_A_l2_sq': B_to_A_l2_sq,
         'A_to_B_l2': A_to_B_l2,
         'B_to_A_l2': B_to_A_l2,
         'n_samples_per_mesh': n_samples,
@@ -148,5 +166,5 @@ if __name__ == "__main__":
     gt_faces = pred_faces.copy()
 
     metrics = chamfer_distance_from_meshes(pred_verts, pred_faces, gt_verts, gt_faces,
-                                           n_samples=20000)
+                                           n_samples=100000)
     print("Chamfer metrics:", metrics)