cleaning

compute_features.py

 
-
-
-import math
 import numpy as np
 import json
 from utils import compute_categories_id, compute_id_good_occ
@@ -51,7 +48,6 @@ def process_compute(data_name, camera, camera_resized, new_size, Nb_camera, Worl
                 with open(f'{data_name}/Generated/Count_{p-1}.json') as f:
                     list_count_categories = json.load(f)
 
-
             for categories in list_categories:
                 if categories in catergories_occ_array.keys():
                     if len(catergories_occ_array[categories]) == 1 :
@@ -73,7 +69,6 @@ def process_compute(data_name, camera, camera_resized, new_size, Nb_camera, Worl
                         meta['occlusion'] = occ_target
                         meta['Nb_instance_category'] = 1
 
-
                         if not os.path.isfile(f'{data_name}/Generated/{categories}/Meta_Gen/{categories}.json'):
                             with open(f'{data_name}/Generated/{categories}/Meta_Gen/{categories}.json', mode='w') as f:
                                 feeds = {}
@@ -86,9 +81,6 @@ def process_compute(data_name, camera, camera_resized, new_size, Nb_camera, Worl
                             with open(f'{data_name}/Generated/{categories}/Meta_Gen/{categories}.json', mode='w') as f:
                                 f.write(json.dumps(feeds, indent=4))
 
-                        # with open(f'{data_name}/Generated/Meta_Gen/{categories}/{categories}.json', "a") as meta_file:
-                        #     json.dump(meta, meta_file, indent=4)
-
                         for k in range(len(data_3D_pose)):
 
                             if data_3D_pose[k]['id'] == catergories_occ_array[categories][0]:
@@ -122,73 +114,44 @@ def process_compute(data_name, camera, camera_resized, new_size, Nb_camera, Worl
 
                         instance_img = instance(img, id)
                         cv2.imwrite(f"{data_name}/Generated/{categories}/Instance_Mask/{p}.png", 255*instance_img)
-                        cat_mask = cv2.resize(instance_img, new_size)
-                        cv2.imwrite(f"{data_name}/Generated/{categories}/Instance_Mask_resized/{p}.png", 255*cat_mask)
+                        instance_img_resized = cv2.resize(instance_img, new_size)
+                        cv2.imwrite(f"{data_name}/Generated/{categories}/Instance_Mask_resized/{p}.png", 255*instance_img_resized)
 
                         img = cv2.imread(f"{data_name}/RGB/{p}.png")
                         cv2.imwrite(f"{data_name}/Generated/{categories}/RGB_Gen/{p}.png", img)
-
                         img_resized = cv2.resize(img, new_size)
                         cv2.imwrite(f"{data_name}/Generated/{categories}/RGB_resized/{p}.png", img_resized)
 
                         np.set_printoptions(precision=15)
                         pose = np.load(f'{data_name}/Generated/{categories}/Pose_transformed/{p}.npy')
-                        #print(pose)
                         R_exp = pose[0:3, 0:3]
                         tVec = pose[0:3, 3]
 
-                        #print(tVec)
-                        # camera = np.matrix([[1386.4138492513919, 0.0, 960.5],
-                        #                     [0.0, 1386.4138492513919, 540.5],
-                        #                     [0.0, 0.0, 1.0]])
-
-                        
                         fps_points = np.loadtxt(f'{data_name}/Generated/{categories}/{categories}_fps_3d.txt')
-                        # process(pcd_bbox, pcd, R_exp, tVec, camera, img)
                         center = fps_points.mean(0)
                         fps_points = np.append(fps_points, [center], axis=0)
-                        
-                        
                         points = process2(fps_points, R_exp, tVec, camera, img, vis)
-                        #out = np.zeros((1, ))
-
-                        out = [int(catergories_occ_array[categories][0])]# [catergories_occ_array[categories][0]] #obj_id #len have to be 1 !!
-                        print(out)
-
-
+                        out = [int(catergories_occ_array[categories][0])] #len have to be 1 !!
                         ind = 1
                         for point in points:
-                            #out[0][ind] = point[0][0] / img.shape[1]
-                            #[0][ind + 1] = point[0][1] / img.shape[0]
                             x = point[0][0] / img.shape[1]
                             y = point[0][1] / img.shape[0]
                             out.append(x)
                             out.append(y)
                             ind += 2
-
                         np.savetxt(f'{data_name}/Generated/{categories}/FPS/{p}.txt',  np.array(out).reshape(1, len(out)))
-                        #print("stop")
 
                         points_resized = process2(fps_points, R_exp, tVec, camera_resized, img_resized, vis)
-                        #out = np.zeros((1, ))
-
-                        out_resized = [int(catergories_occ_array[categories][0])]# [catergories_occ_array[categories][0]] #obj_id #len have to be 1 !!
-                        print(out)
-
-
+                        out_resized = [int(catergories_occ_array[categories][0])] #len have to be 1 !
                         ind_resized = 1
                         for point_resized in points_resized:
-                            #out[0][ind] = point[0][0] / img.shape[1]
-                            #[0][ind + 1] = point[0][1] / img.shape[0]
                             x_resized = point_resized[0][0] / img_resized.shape[1]
                             y_resized = point_resized[0][1] / img_resized.shape[0]
                             out_resized.append(x_resized)
                             out_resized.append(y_resized)
                             ind_resized += 2
-
                         np.savetxt(f'{data_name}/Generated/{categories}/FPS_resized/{p}.txt',  np.array(out_resized).reshape(1, len(out_resized)))
-                        #print("stop")         
-
+        
     with open(f'{data_name}/Generated/Count_{p}.json', mode='w') as f:
         f.write(json.dumps(list_count_categories, indent=4))
     print(list_count_categories)

main.py

 import os
 import numpy as np
-import json
 from prepare_data import reform_data
-#from pose import transform_pose
-#from bbox_2d import generate_2d_bbox
-#from instance_mask import generate_instance_mask
 from fps_alg import apply_fps
 from bbox_3d import get_3D_bbox
 from compute_features import process_compute
-from utils import compute_categories_id, compute_id_good_occ
-import shutil
 import open3d as o3d
-# Import the library
-import argparse
 from scipy.spatial import distance
+import argparse
 
 def generate_folders(name, list_categories):
     is_exist = os.path.exists(name)
@@ -52,31 +45,20 @@ if __name__ == '__main__':
     # Parse the argument
     args = parser.parse_args()
 
-
     ### parameters ###
     Categories = [] # to read
     Nb_instance = 1
     occ_target = 0.5
     dataset_src = "/media/gduret/DATA/dataset/s2rg/Fruits_all_medium/data"
-    #dataset_src = "/media/mahmoud/E/Fruits_easy/data"
     choice = "low" # depth of rgb resolution datas
     data_options = {"high": "ground_truth_rgb",
                     "low": "ground_truth_depth"}
     dataset_type = data_options[choice]
     dataset_name = f"GUIMOD_{choice}"
     list_categories = ["banana1", "kiwi1", "pear2", "strawberry1", "apricot", "orange2", "peach1", "lemon2", "apple2" ]
-    # frame = "1_600000000"
-    #frame = "1_926000000"
     Nb_camera = 15
-    #Nb_world = 2
 
     generate_folders(dataset_name, list_categories)
-    # for cat in list_categories:
-    #     src_bbox = f"Models/{cat}/{cat.lower()}.ply"
-    #     dst_bbox = f"{dataset_name}/Generated/{cat}/{cat.lower()}.ply"
-    #     shutil.copy(src_bbox, dst_bbox)
-
-
 
     if choice == 'high':
         camera = np.matrix([[1386.4138492513919, 0.0, 960.5],
@@ -94,13 +76,11 @@ if __name__ == '__main__':
                        [0.0, 0.0, 1.0]])
 
     new_camera = trans @ camera
-
     np.savetxt(f'{dataset_name}/Generated/camera_{choice}.txt', camera)
 
     reform_data(dataset_src, dataset_name, dataset_type, Nb_camera, args.World_begin, args.Nb_worlds)
 
     list_categories = ["banana1", "kiwi1", "pear2", "strawberry1", "apricot", "orange2", "peach1", "lemon2", "apple2" ]
-
     objs = {"banana1": [ 0.02949700132012367249, 0.1511049866676330566, 0.06059300713241100311 ],
             "kiwi1": [ 0.04908600077033042908, 0.07206099480390548706, 0.04909799993038177490 ],
             "pear2": [ 0.06601099669933319092, 0.1287339925765991211, 0.06739201396703720093 ],
@@ -115,49 +95,17 @@ if __name__ == '__main__':
         point_cloud = f"Models/{categories}/{categories.lower()}.ply"
         pcd = o3d.io.read_point_cloud(point_cloud)
 
-        #print("pcd", pcd)
-
         fps_points = apply_fps(pcd, 8)
-        #print(fps_points)
         np.savetxt(f'{dataset_name}/Generated/{categories}/{categories}_fps_3d.txt', fps_points)
 
-        #point_cloud = f'/home/mahmoud/PycharmProjects/data/GUIMOD_low/Models/{obj}/{obj.lower()}.ply'
-        #pcd = o3d.io.read_point_cloud(point_cloud)
         point_cloud_in_numpy = np.asarray(pcd.points)
         dim = calc_pts_diameter2(point_cloud_in_numpy) * 100
-
-        print(dim)
-
         np.savetxt(f'{dataset_name}/Generated/{categories}/{categories}_diameter.txt', np.array([dim]))
 
         size_bb = objs[categories]
-
         ext = [x / 2 for x in size_bb]
         bbox = get_3D_bbox(ext)
         np.savetxt(f'{dataset_name}/Generated/{categories}/{categories}_bbox_3d.txt', bbox)  # save
 
-        # json_num = 2
-        # catergories_instance_array_id_to_cat, catergories_instance_array_cat_to_id, catergories_label_to_id = compute_categories_id(dataset_name, json_num)
-        # print(f'{dataset_name}/Bbox_3d/{json_num}.json') 
-        # with open(f"{dataset_name}/Bbox_3d/{json_num}.json", 'r') as f:
-        #     data_Bbox_3d = json.load(f)
-
-        # print("catergories_instance_array_cat_to_id : ", catergories_instance_array_cat_to_id)
-        # print("data_Bbox_3d : ", data_Bbox_3d)
-
-        # for k in range(len(data_Bbox_3d)):
-        #     if data_Bbox_3d[k]['id'] in catergories_instance_array_cat_to_id[categories]:
-        #         print(data_Bbox_3d)
-        #         size_bb = data_Bbox_3d[k]["bbox"]["size"]
-        #         ext = [x / 2 for x in size_bb]
-        #         bbox = get_3D_bbox(ext)
-        #         np.savetxt(f'{dataset_name}/Generated/{categories}/{categories}_bbox_3d.txt', bbox)  # save
-        #         break
-
-
     process_compute(dataset_name, camera, new_camera, new_size, Nb_camera, args.World_begin, args.Nb_worlds, list_categories, occ_target, False)
-    #transform_pose(dataset_name, Nb_camera, Nb_world, list_categories, occ_target)
-    #generate_2d_bbox(dataset_name, Nb_camera, Nb_world, list_categories, occ_target)
-    #generate_instance_mask(dataset_name, Nb_camera, Nb_world, list_categories, occ_target)
-    #generate_fps(dataset_name, camera, Nb_camera, Nb_world, list_categories, occ_target, True)
-    #generate_3d_bbox(dataset_name)
+