bpt / utils.py
whaohan's picture
init commit
ada4b81 verified
import trimesh
import numpy as np
from x_transformers.autoregressive_wrapper import top_p, top_k
class Dataset:
'''
A toy dataset for inference
'''
def __init__(self, input_type, input_list):
super().__init__()
self.data = []
if input_type == 'pc_normal':
for input_path in input_list:
# load npy
cur_data = np.load(input_path)
# sample 4096
assert cur_data.shape[0] >= 4096, "input pc_normal should have at least 4096 points"
idx = np.random.choice(cur_data.shape[0], 4096, replace=False)
cur_data = cur_data[idx]
self.data.append({'pc_normal': cur_data, 'uid': input_path.split('/')[-1].split('.')[0]})
elif input_type == 'mesh':
mesh_list, pc_list = [], []
for input_path in input_list:
# sample point cloud and normal from mesh
cur_data = trimesh.load(input_path, force='mesh')
cur_data = apply_normalize(cur_data)
mesh_list.append(cur_data)
pc_list.append(sample_pc(cur_data, pc_num=4096, with_normal=True))
for input_path, cur_data in zip(input_list, pc_list):
self.data.append({'pc_normal': cur_data, 'uid': input_path.split('/')[-1].split('.')[0]})
print(f"dataset total data samples: {len(self.data)}")
def __len__(self):
return len(self.data)
def __getitem__(self, idx):
data_dict = {}
data_dict['pc_normal'] = self.data[idx]['pc_normal']
data_dict['uid'] = self.data[idx]['uid']
return data_dict
def joint_filter(logits, k = 50, p=0.95):
logits = top_k(logits, k = k)
logits = top_p(logits, thres = p)
return logits
def apply_normalize(mesh):
'''
normalize mesh to [-1, 1]
'''
bbox = mesh.bounds
center = (bbox[1] + bbox[0]) / 2
scale = (bbox[1] - bbox[0]).max()
mesh.apply_translation(-center)
mesh.apply_scale(1 / scale * 2 * 0.95)
return mesh
def sample_pc(mesh_path, pc_num, with_normal=False):
mesh = trimesh.load(mesh_path, force='mesh', process=False)
mesh = apply_normalize(mesh)
if not with_normal:
points, _ = mesh.sample(pc_num, return_index=True)
return points
points, face_idx = mesh.sample(50000, return_index=True)
normals = mesh.face_normals[face_idx]
pc_normal = np.concatenate([points, normals], axis=-1, dtype=np.float16)
# random sample point cloud
ind = np.random.choice(pc_normal.shape[0], pc_num, replace=False)
pc_normal = pc_normal[ind]
return pc_normal