File size: 6,058 Bytes
352b049
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import numpy as np
import smplx
import torch

from .optimize.smplify import SMPLify3D

class joints2smpl:
    def __init__(
        self,
        device,
        num_smplify_iters=150,
        SMPL_MODEL_DIR= "./body_models/",
        GMM_MODEL_DIR="./joint2smpl_models/",
    ):
        self.device = torch.device(device)
        # self.batch_size = num_frames
        self.num_joints = 22  # for HumanML3D
        self.joint_category = "AMASS"
        self.num_smplify_iters = num_smplify_iters
        self.fix_foot = False
        smplmodel = smplx.create( SMPL_MODEL_DIR, model_type="smpl", gender="neutral", ext="pkl").to(self.device)
        # smplmodel = smplx.create(
        #      SMPL_MODEL_DIR, model_type="smpl", gender="female", ext="pkl"
        # ).to(self.device)


        # # #-------------initialize SMPLify
        # self.smplify = SMPLify3D(smplxmodel=smplmodel, batch_size=self.batch_size, joints_category=self.joint_category, num_iters=self.num_smplify_iters, device=self.device)
        self.smplify = SMPLify3D(
            smplxmodel=smplmodel,
            # batch_size=self.batch_size,
            joints_category=self.joint_category,
            num_iters=self.num_smplify_iters,
            device=self.device,
            GMM_MODEL_DIR=GMM_MODEL_DIR
        )

    def joint2smpl(self, input_joints, init_params=None, hmp=True, fix_betas=False):
        batch_size = input_joints.shape[0]

        _smplify = self.smplify  # if init_params is None else self.smplify_fast
        pred_pose = torch.zeros(batch_size, 72).to(self.device)
        pred_betas = torch.zeros(1, 10).expand(batch_size, 10).to(self.device)
        pred_cam_t = torch.zeros(1, 3).expand(batch_size, 3).to(self.device)
        keypoints_3d = torch.Tensor(input_joints).to(self.device).float()

        if init_params is None:
            # assert 0, "Not implemented. Missing init pose."
            pred_betas = (
                torch.tensor([ 1.47646511e+00,  4.79749501e-01, -6.36047006e-01, -1.52980864e+00,
                                -1.11884427e+00, -5.40487289e-01,  3.93005997e-01, -1.88832569e+00,
                                -2.78680950e-01, -5.49529344e-02])
                .unsqueeze(0)
                .repeat(batch_size, 1)
                .float()
                .to(self.device)
            )
            pred_pose = (torch.tensor([ 0.4531,  0.3044,  0.2968, -0.2239,  0.0174,  0.0925, -0.2378, -0.0465,
                                    -0.0786,  0.2782,  0.0141,  0.0138,  0.4328, -0.0629, -0.0961,  0.5043,
                                    0.0035,  0.0610,  0.0230, -0.0317,  0.0058,  0.0070,  0.1317, -0.0544,
                                    -0.0589, -0.1752,  0.1355,  0.0134, -0.0037,  0.0089, -0.2093,  0.1600,
                                    0.1092, -0.0387,  0.0824, -0.2041, -0.0056, -0.0075, -0.0035, -0.0237,
                                    -0.1248, -0.2736, -0.0459,  0.1991,  0.2373,  0.0667, -0.0405,  0.0329,
                                    0.0536, -0.2914, -0.6969,  0.0559,  0.2858,  0.6525,  0.1222, -0.9116,
                                    0.2383, -0.0366,  0.9237, -0.2554, -0.0657, -0.1045,  0.0501, -0.0388,
                                    0.0909, -0.0707, -0.1437, -0.0590, -0.1801, -0.0875,  0.1093,  0.2009])
                                    .unsqueeze(0)
                                    .repeat(batch_size, 1)
                                    .float().to(self.device)
                        )   
            pred_cam_t = torch.Tensor([0.0, 0.0, 0.0]).unsqueeze(0).to(self.device)
        else:
            pred_betas = init_params["betas"]
            pred_pose = init_params["pose"]
            pred_cam_t = init_params["cam"]

        if self.joint_category == "AMASS":
            confidence_input = torch.ones(self.num_joints)
            # make sure the foot and ankle
            if self.fix_foot == True:
                confidence_input[7] = 1.5
                confidence_input[8] = 1.5
                confidence_input[10] = 1.5
                confidence_input[11] = 1.5
        else:
            print("Such category not settle down!")

        (
            new_opt_vertices,
            new_opt_joints,
            new_opt_pose,
            new_opt_betas,
            new_opt_cam_t,
        ) = _smplify(
            pred_pose.detach(),
            pred_betas.detach(),
            pred_cam_t.detach(),
            keypoints_3d,
            conf_3d=confidence_input.to(self.device),
            if_simple_hmp_optimizes=hmp,
            fix_betas=fix_betas,
        )

        # thetas = new_opt_pose.reshape(batch_size, 24, 3)
        # thetas = matrix_to_rotation_6d(
        #     axis_angle_to_matrix(thetas)
        # )  # [bs, 24, 6]
        # # root_loc = torch.tensor(keypoints_3d[:, 0])  # [bs, 3]
        # root_loc = keypoints_3d[:, 0].clone()  # [bs, 3]
        # root_loc = torch.cat([root_loc, torch.zeros_like(root_loc)], dim=-1).unsqueeze(
        #     1
        # )  # [bs, 1, 6]
        # thetas = torch.cat([thetas, root_loc], dim=1).permute(1, 2, 0)  # [25, 6, 196]
        # thetas = thetas.clone().detach()

        return new_opt_pose, {
            "pose": new_opt_joints[..., :24, :].clone().detach(),
            "betas": new_opt_betas.clone().detach(),
            "cam": new_opt_cam_t.clone().detach(),
        }


class Skeleton2Obj:
    def __init__(
        self,
        device: str = "cpu",
        num_smplify_iters=150,
        smpl_model_dir="./body_models/",
        gmm_model_dir="./joint2smpl_models/",
    ):
        self.simplify = joints2smpl(
            device=device,
            num_smplify_iters=num_smplify_iters,
            SMPL_MODEL_DIR=smpl_model_dir,
            GMM_MODEL_DIR=gmm_model_dir,
        )

    
    def convert_motion_2smpl(self, motion, hmp=True, init_params=None, fix_betas=False) -> tuple[np.ndarray, np.ndarray]:
        new_opt_rot, smpl_dict = self.simplify.joint2smpl(motion, hmp=hmp, init_params=init_params, fix_betas=fix_betas)
        return new_opt_rot, smpl_dict