MCPcopy Create free account
hub / github.com/MotrixLab/AiOS / forward

Method forward

util/smplx/smplx/body_models.py:1035–1206  ·  view source on GitHub ↗

Forward pass for the SMPLX model. Parameters ---------- global_orient: torch.tensor, optional, shape Bx3 If given, ignore the member variable and use it as the global rotation of the body. Useful if someone wishes to predicts this with an

(self,
                betas: Optional[Tensor] = None,
                global_orient: Optional[Tensor] = None,
                body_pose: Optional[Tensor] = None,
                left_hand_pose: Optional[Tensor] = None,
                right_hand_pose: Optional[Tensor] = None,
                transl: Optional[Tensor] = None,
                expression: Optional[Tensor] = None,
                jaw_pose: Optional[Tensor] = None,
                leye_pose: Optional[Tensor] = None,
                reye_pose: Optional[Tensor] = None,
                return_verts: bool = True,
                return_full_pose: bool = False,
                pose2rot: bool = True,
                **kwargs)

Source from the content-addressed store, hash-verified

1033 return '\n'.join(msg)
1034
1035 def forward(self,
1036 betas: Optional[Tensor] = None,
1037 global_orient: Optional[Tensor] = None,
1038 body_pose: Optional[Tensor] = None,
1039 left_hand_pose: Optional[Tensor] = None,
1040 right_hand_pose: Optional[Tensor] = None,
1041 transl: Optional[Tensor] = None,
1042 expression: Optional[Tensor] = None,
1043 jaw_pose: Optional[Tensor] = None,
1044 leye_pose: Optional[Tensor] = None,
1045 reye_pose: Optional[Tensor] = None,
1046 return_verts: bool = True,
1047 return_full_pose: bool = False,
1048 pose2rot: bool = True,
1049 **kwargs) -> SMPLXOutput:
1050 """Forward pass for the SMPLX model.
1051
1052 Parameters
1053 ----------
1054 global_orient: torch.tensor, optional, shape Bx3
1055 If given, ignore the member variable and use it as the global
1056 rotation of the body. Useful if someone wishes to predicts this
1057 with an external model. (default=None)
1058 betas: torch.tensor, optional, shape Bx10
1059 If given, ignore the member variable `betas` and use it
1060 instead. For example, it can used if shape parameters
1061 `betas` are predicted from some external model.
1062 (default=None)
1063 expression: torch.tensor, optional, shape Bx10
1064 If given, ignore the member variable `expression` and use it
1065 instead. For example, it can used if expression parameters
1066 `expression` are predicted from some external model.
1067 body_pose: torch.tensor, optional, shape Bx(J*3)
1068 If given, ignore the member variable `body_pose` and use it
1069 instead. For example, it can used if someone predicts the
1070 pose of the body joints are predicted from some external model.
1071 It should be a tensor that contains joint rotations in
1072 axis-angle format. (default=None)
1073 left_hand_pose: torch.tensor, optional, shape BxP
1074 If given, ignore the member variable `left_hand_pose` and
1075 use this instead. It should either contain PCA coefficients or
1076 joint rotations in axis-angle format.
1077 right_hand_pose: torch.tensor, optional, shape BxP
1078 If given, ignore the member variable `right_hand_pose` and
1079 use this instead. It should either contain PCA coefficients or
1080 joint rotations in axis-angle format.
1081 jaw_pose: torch.tensor, optional, shape Bx3
1082 If given, ignore the member variable `jaw_pose` and
1083 use this instead. It should either joint rotations in
1084 axis-angle format.
1085 transl: torch.tensor, optional, shape Bx3
1086 If given, ignore the member variable `transl` and use it
1087 instead. For example, it can used if the translation
1088 `transl` is predicted from some external model.
1089 (default=None)
1090 return_verts: bool, optional
1091 Return the vertices. (default=True)
1092 return_full_pose: bool, optional

Callers

nothing calls this directly

Calls 4

lbsFunction · 0.90
vertices2landmarksFunction · 0.90
SMPLXOutputClass · 0.85

Tested by

no test coverage detected