from torch.utils.data import Dataset import numpy as np from human_corres.utils import helper from human_corres.data import Data import torch_geometric import torch_geometric.transforms as T import human_corres as hc import human_corres.transforms as H num_test = 5000 TrainTransform = T.Compose([ T.Center(), T.RandomRotate(30, axis=0), T.RandomRotate(30, axis=1), T.RandomRotate(30, axis=2), H.GridSampling(0.01) ]) TestTransform = T.Compose([ T.Center(), H.GridSampling(0.01) ]) class SurrealFEPts(Dataset): """Surreal 3D points for Feature Extraction (FE). Output: dictionary with keys {points3d, correspondence} Data Format: points3d: [num_points, 3] real numbers. correspondence: [num_points] integers in range [6890]. """ def __init__(self, descriptor_dim, split='train', desc='Laplacian_n',
import torch from torch.utils.data import Dataset import numpy as np from human_corres.utils import helper from human_corres.data import Data import human_corres as hc import scipy.io as sio import torch_geometric.transforms as T import human_corres.transforms as H TrainTransform = T.Compose([ T.Center(), T.RandomRotate(30, axis=0), T.RandomRotate(30, axis=1), T.RandomRotate(30, axis=2), H.GridSampling(0.01), ]) TestTransform = T.Compose([ T.Center(), H.GridSampling(0.01), ]) class SMALFEPts(Dataset): """SMAL 3D points for Feature Extraction (FE). Output: dictionary with keys {points3d, correspondence} Data Format: points3d: [num_points, 3] real numbers. correspondence: [num_points] integers in range [6890].
from human_corres.utils import helper from human_corres.data import Data import torch_geometric import torch_geometric.transforms as T import human_corres as hc import human_corres.transforms as H num_views = 20 IDlist = np.arange(5000 * num_views) num_test = 50 TrainTransform = T.Compose([ T.Center(), T.RandomRotate(30, axis=0), T.RandomRotate(30, axis=1), T.RandomRotate(30, axis=2), H.GridSampling(0.01) ]) TestTransform = T.Compose([T.Center(), H.GridSampling(0.01)]) class DGFSurrealFEPts(Dataset): """Surreal 3D points for Feature Extraction (FE). Output: dictionary with keys {points3d, correspondence} Data Format: points3d: [num_points, 3] real numbers. correspondence: [num_points] integers in range [6890]. """ def __init__(self, descriptor_dim, split='train',