-
Notifications
You must be signed in to change notification settings - Fork 4
/
demo_smplify_dc.py
183 lines (154 loc) · 7.73 KB
/
demo_smplify_dc.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
# -*- coding: utf-8 -*-
# Max-Planck-Gesellschaft zur Förderung der Wissenschaften e.V. (MPG) is
# holder of all proprietary rights on this computer program.
# You can only use this computer program if you have closed
# a license agreement with MPG or you get the right to use the computer
# program from someone who is authorized to grant you that right.
# Any use of the computer program without a valid license is prohibited and
# liable to prosecution.
#
# Copyright©2021 Max-Planck-Gesellschaft zur Förderung
# der Wissenschaften e.V. (MPG). acting on behalf of its Max Planck Institute
# for Intelligent Systems. All rights reserved.
#
# Contact: [email protected]
import os
import os.path as osp
import cv2
# hack to get the correct gpu device id on cluster
os.environ['PYOPENGL_PLATFORM'] = 'egl'
import numpy as np
import torch
import pickle
from tqdm import tqdm
from torchgeometry import rotation_matrix_to_angle_axis
from configs.smplify_dc_options import SMPLifyDCOptions
from configs import config
from tuch.models.hmr import hmr
from tuch.smplify.smplifydc import SMPLifyDC
from data.essentials import constants
from tuch.datasets.base_dataset import BaseDataset
from tuch.models.smpl import SMPL
from tuch.utils.renderer import Renderer
from data.essentials.segments.smpl import segm_utils as exn
from tuch.utils.segmentation import BatchBodySegment
def main(options):
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
batch_size = options.batch_size
dataset = BaseDataset(options, options.ds_names[0], use_augmentation=False)
# Load SPIN to initialize the optimization
modelspin = hmr(config.SMPL_MEAN_PARAMS).to(device)
spincheckpoint = torch.load(config.SPIN_MODEL_CHECKPOINT)
modelspin.load_state_dict(spincheckpoint['model'], strict=False)
modelspin.eval()
# load body model
smpl = SMPL(config.SMPL_MODEL_DIR,
batch_size=options.batch_size,
create_transl=False
).to(device)
face_tensor = torch.tensor(smpl.faces.astype(np.int64),
dtype=torch.long, device=device) \
.unsqueeze_(0) \
.expand(options.batch_size,-1,-1)
geodistssmpl = torch.tensor(np.load(config.GEODESICS_SMPL),
device=device)
# load optimization routine
smplify = SMPLifyDC(step_size=1e-2,
batch_size=options.batch_size,
num_iters=options.num_smplify_iters,
focal_length=constants.FOCAL_LENGTH,
geodistssmpl=geodistssmpl,
geothres=config.geothres,
)
# load dsc data
classes = pickle.load(open(osp.join(config.DSC_ROOT, 'classes.pkl'), 'rb'))
csig = pickle.load(open(osp.join(config.DSC_ROOT, 'ContactSigSMPL.pkl'), 'rb'))
contactlist = {'classes': classes, 'csig': csig}
has_smpl_= torch.zeros((options.batch_size)).to(device).bool()
# Setup renderer for visualization
renderer = Renderer(focal_length=constants.FOCAL_LENGTH,
img_res=constants.IMG_RES,
faces=smpl.faces,
contactlist=contactlist)
# segments
segments = BatchBodySegment([x for x in exn.segments.keys()], face_tensor[0])
# Process each image
for idx in tqdm(range(len(dataset.data['imgname']))):
batch = dataset[idx]
# create tensor (this is what the torch data loader normally does)
batch = {k: torch.tensor(v).unsqueeze(0).to(device) if not isinstance(v, str) \
else [v] for k,v in batch.items()}
# move input to device
batch = {k: v.to(device) if isinstance(v, torch.Tensor) \
else v for k,v in batch.items()}
images = batch['img']
gt_keypoints_2d = batch['keypoints']
# De-normalize 2D keypoints from [-1,1] to pixel space
gt_keypoints_2d_orig = gt_keypoints_2d.clone()
gt_keypoints_2d_orig[:, :, :-1] = 0.5 * options.img_res * \
(gt_keypoints_2d_orig[:, :, :-1] + 1)
gt_disc_contact = batch['contact_vec']
has_disc_contact = batch['has_disc_contact'].bool()
has_2d_keypoints_gtanno= batch['has_gt_kpts'].bool()
# Get the fits of the original SPIN model. To add camera loss
with torch.no_grad():
# forward pass SPIN model in eval mode
init_rotmat, init_betas, init_camera = modelspin(images)
output = smpl(betas=init_betas, body_pose=init_rotmat[:,1:],
global_orient=init_rotmat[:,0].unsqueeze(1), pose2rot=False)
init_vertices = output.vertices.clone()
init_joints = output.joints.clone()
init_cam_t = torch.stack([init_camera[:,1],
init_camera[:,2],
2*constants.FOCAL_LENGTH/(options.img_res * \
init_camera[:,0] + 1e-9)],dim=-1)
# Convert predicted rotation matrices to axis-angle
init_rotmat_hom = torch.cat([init_rotmat.detach().view(-1, 3, 3).detach(),
torch.tensor([0,0,1], dtype=torch.float32, device=device) \
.view(1, 3, 1).expand(batch_size * 24, -1, -1)], dim=-1)
init_pose = rotation_matrix_to_angle_axis(init_rotmat_hom).contiguous().view(batch_size, -1)
init_pose[torch.isnan(init_pose)] = 0.0
new_opt_vertices, new_opt_joints,\
new_opt_pose, new_opt_betas,\
new_opt_cam_t, new_opt_joint_loss, \
smplifyoptiverts = smplify(
init_pose.detach(),
init_betas.detach(),
init_cam_t.detach(),
0.5 * options.img_res * \
torch.ones(batch_size, 2, device=device),
gt_keypoints_2d_orig,
use_contact=options.use_contact_in_the_loop,
contactlist=contactlist,
gt_contact=[gt_disc_contact, None],
ignore_idxs=has_smpl_,
has_discrete_contact=has_disc_contact,
has_gt_keypoints=has_2d_keypoints_gtanno,
contact_loss_weight=options.contact_in_the_loop_loss_weight,
contact_loss_return='sum',
segments=segments)
# render results to logdir
images = images * torch.tensor([0.229, 0.224, 0.225], device=images.device) \
.reshape(1,3,1,1)
images = images + torch.tensor([0.485, 0.456, 0.406], device=images.device) \
.reshape(1,3,1,1)
aroundy = cv2.Rodrigues(np.array([0, np.radians(90.), 0]))[0]
for idx in range(batch_size):
imgname, imgending = batch['imgname'][idx].split('/')[-1].split('.')
img = images[idx].permute(1,2,0).cpu().numpy()
img_out = images[idx].permute(1,2,0).cpu().numpy()
for data in [(init_vertices[idx], init_cam_t[idx]),
(new_opt_vertices[idx],new_opt_cam_t[idx])]:
verts, cam = data[0].cpu().numpy(), data[1].cpu().numpy()
img_out_front = renderer(verts, cam, img,
contact=gt_disc_contact[idx])
center = verts.mean(axis=0)
rot_vertices = np.dot((verts - center), aroundy) + center
img_out_rot = renderer(rot_vertices, cam, np.zeros_like(img),
contact=gt_disc_contact[idx])
img_out = np.hstack((img_out, img_out_front, img_out_rot))
cv2.imwrite(osp.join(options.log_dir, '.'.join([imgname, imgending])),
img_out[:,:,::-1]*255)
if __name__ == '__main__':
options = SMPLifyDCOptions().parse_args()
main(options)