Functions — github.com/JubSteven/POEM-v2

CodeHub MCPcopy Create free account

hub / github.com/JubSteven/POEM-v2 / functions

Functions1,165 in github.com/JubSteven/POEM-v2

⨍ Functions 1,165◇ Types & classes 149

Functioncollation_random_n_views

lib/utils/collation.py:7

Functioncombine_view

(view_list, ncol=None)

lib/utils/vis_cv2_util.py:26

Methodcompute_loss

lib/models/model_abc.py:41

Methodcondition

lib/fit/frame_fit/one_frame_fit_silh.py:255

Methodcondition

lib/fit/frame_fit/one_frame_fit.py:287

Methodcondition

lib/viztools/viz_o3d_utils.py:204

Functionconstant_init

(module, val, bias=0)

lib/utils/net_utils.py:157

Functionconv1x1

1x1 convolution

lib/models/backbones/resnet.py:26

lib/external/metro/data_adaptor.py:23

Functioncopysign

Return a tensor where each element has the absolute value taken from the, corresponding element of a, with sign taken from the corresponding

transform/rotation_jit.py:9

Functioncopysign_np

Return a ndarray where each element has the absolute value taken from the, corresponding element of a, with sign taken from the corresponding

transform/rotation_np.py:5

Functioncreate_coord_system_can

(scale=1, transf=None, o3d=None)

lib/viztools/viz_o3d_utils.py:228

(self, msg, **kwargs)

lib/utils/logger.py:109

Functioncvt_from_trimesh

(mesh: trimesh.Trimesh, o3d=None)

lib/viztools/viz_o3d_utils.py:214

Functiondecaption_pos

lib/utils/vis_cv2_util.py:394

Functiondraw_wireframe_hand

(img, hand_joint_arr, hand_joint_mask)

lib/utils/vis_cv2_util.py:282

Functiondraw_wireframe_hand_kp

(img, hand_joint_arr, hand_joint_mask)

lib/utils/vis_cv2_util.py:323

Functiondraw_wireframe_hand_kp_large

(img, hand_joint_arr, hand_joint_mask)

lib/utils/vis_cv2_util.py:335

Functiondraw_wireframe_hand_large

(img, hand_joint_arr, hand_joint_mask)

lib/utils/vis_cv2_util.py:294

Functiondraw_wireframe_markerset_body

(img, body_markerset_arr, body_markerset_mask)

lib/utils/vis_cv2_util.py:529

Functiondraw_wireframe_markerset_hand

(img, hand_markerset_arr, hand_markerset_mask)

lib/utils/vis_cv2_util.py:541

Functionenable_lower_param

lib/utils/misc.py:30

Functioneuler_angle_to_quat

Convert rotations given as Euler angles in radians to quaternions. Args: euler_angles: Euler angles in radians as tensor of shape (..

transform/rotation_jit.py:360

Functioneuler_angle_to_quat_np

Convert rotations given as Euler angles in radians to quaternions. Args: euler_angles: Euler angles in radians as tensor of shape (..

transform/rotation_np.py:342

Functioneuler_angle_to_rotvec_np

Convert rotations given as Euler angles in radians to axis/angle. Args: euler_angles: Euler angles in radians as tensor of shape (...

transform/rotation_np.py:355

(self, **kwargs)

lib/metrics/basic_metric.py:25

(self, losses, batch_size: int = 1, **kwargs)

lib/metrics/basic_metric.py:74

(self, pred_kp, gt_kp, kp_vis=None, **kwargs)

lib/metrics/mean_epe.py:23

Used to feed data to the class. Stores the euclidean distance between gt and pred, when it is visible.

lib/metrics/pck.py:37

Used to feed data to the class. Stores the euclidean distance between gt and pred, when it is visible.

lib/datasets/ho3d_official_test.py:330

Methodfeed_forward_chunk

(self, attention_output)

lib/models/bricks/pt_metro_transformer.py:88

Methodformat_metric

(self, mode="train")

lib/models/PETR.py:411

Methodformat_metric

(self, mode="train")

lib/models/darkpose.py:316

Methodformat_metric

(self, mode="train")

lib/models/POEM.py:672

Methodformat_metric

(self, mode="train")

lib/external/metro/metro.py:202

mode is in ['persp','ortho']

lib/utils/neural_renderer.py:16

Args: cam_intr: camera intrinsics [NPERSP, 3, 3] cam_extr: camera extrinsics [NPERSP, 4, 4] hand_verts: [

lib/fit/silhouette_loss.py:29

( self, vertices, faces, intr, R=None, t=None, dist_co

lib/fit/pytorch3d_renderer.py:45

(self, inputs, step_idx, mode="train", **kwargs)

lib/models/PETR.py:424

(self, output, target)

lib/models/darkpose.py:54

(self, inputs, step_idx, mode="train", **kwargs)

lib/models/darkpose.py:328

(self, inputs, step_idx, mode="train", **kwargs)

lib/models/integal_pose.py:55

(self, **kwargs)

lib/models/integal_pose.py:358

(self, inputs, step_idx, mode="train", **kwargs)

lib/models/POEM.py:704

lib/models/backbones/hrnet.py:51

lib/models/backbones/hrnet.py:85

lib/models/backbones/hrnet.py:217

lib/models/backbones/hrnet.py:385

initialized to perform identity transformation. The affine transform `x * weight + bias` will perform the equivalent computat

lib/models/backbones/resnet.py:59

(self, x: torch.Tensor)

lib/models/backbones/resnet.py:90

(self, x: torch.Tensor)

lib/models/backbones/resnet.py:124

(self, **kwargs)

lib/models/backbones/resnet.py:206

lib/models/backbones/hourglass.py:29

lib/models/backbones/hourglass.py:110

Forward function for `TransformerDecoder`. Args: query (Tensor): Input query with shape `(num_query, bs, embed_dim

lib/models/layers/petr_transformer.py:139

Forward function for `TransformerCoder`. Returns: Tensor: forwarded results with shape [num_query, bs, embed_dims].

lib/models/layers/petr_transformer.py:235

Forward function for `MultiheadAttention`. **kwargs allow passing a more general data flow when combining with other operations in `tr

lib/models/layers/petr_transformer.py:315

Forward function for `SinePositionalEncoding`. Args: mask (Tensor): ByteTensor mask. Non-zero values representing

lib/models/layers/petr_transformer.py:434

(ctx, value, value_spatial_shapes, value_level_start_index, sampling_locations, attention_weights,

lib/models/layers/mvp_decoder.py:69

:param query (n_views, Length_{query}, C) :param reference_points (n_views, Length_{query}, n_levels, 2

lib/models/layers/mvp_decoder.py:165

(self, tgt, query_pos, reference_points, src_v

lib/models/layers/mvp_decoder.py:342

( self, tgt, reference_points, src_views, src_views_with_rayembed,

lib/models/layers/mvp_decoder.py:458

(self, pt_xyz, pt_feats, query_xyz, reg_branch

lib/models/layers/ptEmb_transformer.py:66

(self, query_feats, query_xyz, pt_feats, pt_xyz)

lib/models/layers/ptEmb_transformer.py:115

(self, query_xyz, query_feat, point_xyz, point_feat)

lib/models/layers/ptEmb_transformer.py:177

(self, pt_xyz, pt_feats, reg_branches, pt_embe

lib/models/layers/ptEmb_transformer.py:245

(self, query_xyz, query_feat, pt_xyz, pt_feats)

lib/models/layers/ptEmb_transformer.py:371

(self, pose_coeffs, betas=None, **kwargs)

lib/models/layers/mano_wrapper.py:24

Forward function for `MultiheadAttention`. **kwargs allow passing a more general data flow when combining with other operations in `t

lib/models/bricks/transformer.py:87

Forward function for `FFN`. The function would add x to the output tensor if residue is None.

lib/models/bricks/transformer.py:226

Forward function for `TransformerDecoderLayer`. **kwargs contains some specific arguments of attentions. Args: query (Te

lib/models/bricks/transformer.py:345

Forward function for `TransformerCoder`. Args: query (Tensor): Input query with shape `(num_queries, bs, embed_di

lib/models/bricks/transformer.py:474

lib/models/bricks/point_transformers.py:41

(self, xyz, features)

lib/models/bricks/point_transformers.py:61

(self, xyz, features, query)

lib/models/bricks/point_transformers.py:117

lib/models/bricks/metro_transformer.py:19

(self, query_feats, key_feats, query_input_ids=None,

lib/models/bricks/metro_transformer.py:63

# self.bert has three outputs # predictions[0]: output tokens # predictions[1]: all_hidden_states, if enable "self.config.out

lib/models/bricks/metro_transformer.py:169

lib/models/bricks/conv.py:37

(self, pt_xyz, pt_feats, query_xyz, query_feat)

lib/models/bricks/pt_metro_transformer.py:34

(self, query_feats, attention_mask, pt_feats, query_xyz, pt_xyz)

lib/models/bricks/pt_metro_transformer.py:56

(self, query_xyz, query_feats, pt_xyz, pt_feats, head_mask=None)

lib/models/bricks/pt_metro_transformer.py:153

(self, mlvl_feat, img_metas, reference_points, template_mesh, **kwargs)

lib/models/heads/ptEmb_head.py:193

mlvl_feat: multi-view image features reference_points: reference mesh [X, V] where X -> V template_mesh: a flat M

lib/models/heads/ptEmb_head.py:216

(self, mlvl_feat, img_metas, reference_points, template_mesh, **kwargs)

lib/models/heads/ptEmb_head.py:442

(self, mlvl_feat, img_metas, reference_joints, **kwargs)

lib/models/heads/ptEmb_head.py:825

img_feats for ResNet 34: torch.Size([BN, 64, 64, 64]) torch.Size([BN, 128, 32, 32]) torch.Size([BN, 2

lib/models/heads/mvp_head.py:223

(self, mlvl_feats, img_metas, template_mesh, **kwargs)

lib/models/heads/petr_FTL_head.py:74

(self, mlvl_feats, img_metas, template_mesh, **kwargs)

lib/models/heads/petr_head.py:166

lib/external/cmr/net.py:39

:param x: input feature :param up_transform: upsample matrix :return: upsampled and convoluted feature

lib/external/cmr/net.py:74

lib/external/cmr/net.py:115

lib/external/cmr/net.py:156

lib/external/cmr/net.py:183

lib/external/cmr/model.py:38

lib/external/cmr/model.py:63

(self, inputs, step_idx, mode="train", **kwargs)

lib/external/cmr/model.py:505

lib/external/metro/base_model.py:30

(self, img_feats, input_ids=None, token_type_ids=None,

lib/external/metro/base_model.py:69

# self.bert has three outputs # predictions[0]: output tokens # predictions[1]: all_hidden_states, if enable "self.config.out

lib/external/metro/base_model.py:152

← previous next →801–900 of 1,165, ranked by callers