| 47 | |
| 48 | |
| 49 | class ModelNetDataLoader(Dataset): |
| 50 | def __init__(self, root, args, split='train', process_data=False): |
| 51 | self.root = root |
| 52 | self.npoints = args.num_point |
| 53 | self.process_data = process_data |
| 54 | self.uniform = args.use_uniform_sample |
| 55 | self.use_normals = args.use_normals |
| 56 | self.num_category = args.num_category |
| 57 | |
| 58 | if self.num_category == 10: |
| 59 | self.catfile = os.path.join(self.root, 'modelnet10_shape_names.txt') |
| 60 | else: |
| 61 | self.catfile = os.path.join(self.root, 'modelnet40_shape_names.txt') |
| 62 | |
| 63 | self.cat = [line.rstrip() for line in open(self.catfile)] |
| 64 | self.classes = dict(zip(self.cat, range(len(self.cat)))) |
| 65 | |
| 66 | shape_ids = {} |
| 67 | if self.num_category == 10: |
| 68 | shape_ids['train'] = [line.rstrip() for line in open(os.path.join(self.root, 'modelnet10_train.txt'))] |
| 69 | shape_ids['test'] = [line.rstrip() for line in open(os.path.join(self.root, 'modelnet10_test.txt'))] |
| 70 | else: |
| 71 | shape_ids['train'] = [line.rstrip() for line in open(os.path.join(self.root, 'modelnet40_train.txt'))] |
| 72 | shape_ids['test'] = [line.rstrip() for line in open(os.path.join(self.root, 'modelnet40_test.txt'))] |
| 73 | |
| 74 | assert (split == 'train' or split == 'test') |
| 75 | shape_names = ['_'.join(x.split('_')[0:-1]) for x in shape_ids[split]] |
| 76 | self.datapath = [(shape_names[i], os.path.join(self.root, shape_names[i], shape_ids[split][i]) + '.txt') for i |
| 77 | in range(len(shape_ids[split]))] |
| 78 | print('The size of %s data is %d' % (split, len(self.datapath))) |
| 79 | |
| 80 | if self.uniform: |
| 81 | self.save_path = os.path.join(root, 'modelnet%d_%s_%dpts_fps.dat' % (self.num_category, split, self.npoints)) |
| 82 | else: |
| 83 | self.save_path = os.path.join(root, 'modelnet%d_%s_%dpts.dat' % (self.num_category, split, self.npoints)) |
| 84 | |
| 85 | if self.process_data: |
| 86 | if not os.path.exists(self.save_path): |
| 87 | print('Processing data %s (only running in the first time)...' % self.save_path) |
| 88 | self.list_of_points = [None] * len(self.datapath) |
| 89 | self.list_of_labels = [None] * len(self.datapath) |
| 90 | |
| 91 | for index in tqdm(range(len(self.datapath)), total=len(self.datapath)): |
| 92 | fn = self.datapath[index] |
| 93 | cls = self.classes[self.datapath[index][0]] |
| 94 | cls = np.array([cls]).astype(np.int32) |
| 95 | point_set = np.loadtxt(fn[1], delimiter=',').astype(np.float32) |
| 96 | |
| 97 | if self.uniform: |
| 98 | point_set = farthest_point_sample(point_set, self.npoints) |
| 99 | else: |
| 100 | point_set = point_set[0:self.npoints, :] |
| 101 | |
| 102 | self.list_of_points[index] = point_set |
| 103 | self.list_of_labels[index] = cls |
| 104 | |
| 105 | with open(self.save_path, 'wb') as f: |
| 106 | pickle.dump([self.list_of_points, self.list_of_labels], f) |
no outgoing calls
no test coverage detected