Functions — github.com/XunshanMan/MVGFormer

CodeHub MCPcopy Create free account

hub / github.com/XunshanMan/MVGFormer / functions

Functions561 in github.com/XunshanMan/MVGFormer

⨍ Functions 561◇ Types & classes 78

↓ 1 callersMethodloss_poses

Compute the losses related to the bounding poses, the L1 regression loss and the GIoU loss targets dicts must contain the key "pose

lib/models/multi_view_pose_transformer.py:653

↓ 1 callersFunctionmain

process/extract_dataset.py:454

↓ 1 callersFunctionmain

process/extract_number.py:58

↓ 1 callersFunctionmain

run/train_3d.py:152

↓ 1 callersFunctionmain

run/validate_3d.py:95

↓ 1 callersFunctionmask_to_bbox

lib/mvn/datasets/human36m_preprocessing/collect-bboxes.py:33

↓ 1 callersMethodmatch_ref_points_to_gt

OUTPUT: [batch 0: query_ids, gt_ids. batch 1... ... batch N...]

lib/models/dq_decoder.py:1055

↓ 1 callersMethodmerged_log_likelihood

(self, pose, betas)

lib/smpl/prior.py:195

↓ 1 callersMethodmultiple_to_one_matching

sizes: gt_num in each batch. return: [(i,j)...]

lib/models/matcher.py:201

↓ 1 callersMethodmultiple_to_one_matching_KNN

sizes: gt_num in each batch. return: [(i,j)...]

lib/models/matcher.py:232

↓ 1 callersFunctionnested_tensor_from_tensor_list

(tensor_list: List[Tensor])

lib/utils/misc.py:312

↓ 1 callersFunctionnested_tensor_from_tensor_list

(tensor_list: List[Tensor])

lib/models/util/misc.py:399

↓ 1 callersMethodnorm2absolute

(self, norm_coords)

lib/models/mvp_decoder.py:100

↓ 1 callersMethodnorm2absolute

(self, norm_coords)

lib/models/mvp_decoder.py:293

↓ 1 callersFunctionnormalize_image

Normalizes image using ImageNet mean and std Args: image numpy array of shape (h, w, 3): image Returns normalized_image numpy array

lib/mvn/utils/img.py:102

↓ 1 callersMethodpadding_query_with_mask

(self, batch_ids, query_ids, batch_size, mask_ids=0)

lib/models/dq_decoder.py:615

↓ 1 callersFunctionparse_args

lib/dataset/h36m.py:339

↓ 1 callersFunctionparse_args

process/extract_dataset.py:85

↓ 1 callersFunctionparse_args

run/train_3d.py:84

↓ 1 callersFunctionparse_args

run/validate_3d.py:65

↓ 1 callersFunctionparse_value

lib/utils/string_parser.py:20

↓ 1 callersMethodpose_dist

(self, x1, x2, dist='per_joint_mean')

lib/models/matcher.py:61

↓ 1 callersFunctionprecision

Computes the precision@k for the specified values of k

lib/models/multi_view_pose_transformer.py:945

↓ 1 callersFunctionprojectPoints

Projects points X (3xN) using camera intrinsics K (3x3), extrinsics (R,t) and distortion parameters Kd=[k1,k2,p1,p2,k3]. Roughly, x = K*(

lib/utils/transforms.py:184

↓ 1 callersFunctionproject_point_radial

Args x: Nx3 points in world coordinates R: 3x3 Camera rotation matrix T: 3x1 Camera translation parameters f: 2x1

lib/utils/cameras.py:137

↓ 1 callersFunctionproject_point_radial

Args x: Nx3 points in world coordinates R: 3x3 Camera rotation matrix T: 3x1 Camera translation parameters f: (sc

lib/utils/cameras_cpu.py:36

↓ 1 callersFunctionproject_point_radial_batch

Args x: Nx3 points in world coordinates R: 3x3 Camera rotation matrix T: 3x1 Camera translation parameters f: 2x1

lib/utils/cameras.py:167

↓ 1 callersFunctionprojection_transform_batch

(points, center, scale, img_size, device)

lib/utils/cameras.py:28

↓ 1 callersFunctionquat_to_rotmat

Convert quaternion coefficients to rotation matrix. Args: quat: size = [B, 4] 4 <===>(w, x, y, z) Returns: Rotation matrix cor

lib/smpl/utils.py:38

↓ 1 callersFunctionquaternion_to_angle_axis

Convert quaternion vector to angle axis of rotation. Adapted from ceres C++ library: ceres-solver/include/ceres/rotation.h Args: quate

lib/smpl/utils.py:231

↓ 1 callersFunctionreduce_loss

Reduce loss as specified. Args: loss (Tensor): Elementwise loss tensor. reduction (str): Options are "none", "mean" and "sum".

lib/core/loss.py:355

↓ 1 callersMethodreset

lib/core/function.py:594

↓ 1 callersMethodreset

lib/mvn/utils/misc.py:18

↓ 1 callersMethodreset

lib/models/dq_transformer.py:72

↓ 1 callersMethodreset

lib/models/dq_decoder.py:76

↓ 1 callersMethodreset_parameters

lib/models/position_encoding.py:157

↓ 1 callersFunctionrotation_matrix_to_quaternion

Convert 3x4 rotation matrix to 4d quaternion vector This algorithm is based on algorithm described in https://github.com/KieranWynn/pyquaterni

lib/smpl/utils.py:148

↓ 1 callersFunctionsave_config

(config, config_wdb)

run/train_3d.py:148

↓ 1 callersFunctionsave_multiple_images

(images, prefix, normalize=True)

lib/utils/vis.py:338

↓ 1 callersFunctionscale_bbox

lib/mvn/utils/img.py:56

↓ 1 callersFunctionsetup_for_distributed

This function disables printing when not in master process

lib/utils/misc.py:356

↓ 1 callersFunctionsetup_for_distributed

This function disables printing when not in master process

lib/models/util/misc.py:448

↓ 1 callersFunctionsigmoid_focal_loss

Loss used in RetinaNet for dense detection: https://arxiv.org/abs/1708.02002. Args: inputs: A float tensor of arbitrary shape.

lib/models/multi_view_pose_transformer.py:49

↓ 1 callersFunctionto_torch

lib/mvn/utils/img.py:81

↓ 1 callersFunctiontrain_3d

(config, model, optimizer, loader, epoch, output_dir, device=torch.device('cuda'), num_views=5)

lib/core/function.py:56

↓ 1 callersFunctiontriangulate_point_from_multiple_views_linear_torch

Similar as triangulate_point_from_multiple_views_linear() but for PyTorch. For more information see its documentation. Args: proj_matr

lib/mvn/utils/multiview.py:141

↓ 1 callersFunctiontriangulate_point_from_multiple_views_linear_torch_batch

Similar as triangulate_point_from_multiple_views_linear() but for PyTorch. For more information see its documentation. Args: proj_matr

lib/mvn/utils/multiview.py:170

↓ 1 callersMethodtriangulate_ransac

(self, proj_matricies, points, n_iters=10, reprojection_error_epsilon=15, direct_optimization=True)

lib/mvn/models/triangulation.py:72

↓ 1 callersFunctionundistort

(X, meta, iter_num=5)

lib/models/dq_decoder.py:119

↓ 1 callersFunctionunfold_camera_param

lib/utils/cameras_cpu.py:24

↓ 1 callersFunctionunfold_camera_param_batch

(camera, device=None)

lib/utils/cameras.py:118

↓ 1 callersFunctionunit_test_st

(loader, save_dir, prefix='unit_test_st', sigma = 50)

process/extract_dataset.py:209

↓ 1 callersFunctionunit_test_st_least_square

(loader, save_dir, prefix='unit_test_st_least_square')

process/extract_dataset.py:324

↓ 1 callersMethodupdate_after_crop

lib/mvn/utils/multiview.py:23

↓ 1 callersMethodupdate_feature

Update query features with aggregated features from different views.

lib/models/dq_decoder.py:763

↓ 1 callersFunctionuv_to_image_frame

:param uv: (2, N) :param f: scalar :param c: (2, 1) :param k: :param p: :return:

lib/utils/cameras_cuda_col.py:91

↓ 1 callersMethodvisualize_func

A visualization utils function to output mid results. Visualization includes: each input view, projected points, offsets, re

lib/models/dq_decoder.py:719

↓ 1 callersFunctionvisualize_proj_attention

( nviews, rgb_views, src_views, attn_feature_views,

lib/utils/vis.py:82

FunctionPYBIND11_MODULE

lib/models/ops/src/vision.cpp:24

FunctionPose3D_inference

The main procedure of structural triangulation & step contraint algorithm input params: : n_cams : <int> the number of cameras :

lib/structural/structural_triangulation.py:412

(self, images, vertices, translation)

lib/smpl/renderer.py:43

Method__getattr__

lib/utils/misc.py:198

Method__getattr__

lib/models/util/misc.py:284

Method__getitem__

lib/dataset/shelf.py:244

Method__getitem__

lib/dataset/campus.py:239

Method__getitem__

lib/dataset/JointsDataset.py:85

Method__getitem__

lib/mvn/datasets/human36m.py:116

(self, window_size=20, fmt=None)

lib/utils/misc.py:55

(self, delimiter="\t")

lib/utils/misc.py:187

(self, tensors, mask: Optional[Tensor])

lib/utils/misc.py:334

(self, cfg, image_set, is_train, transform=None)

lib/dataset/shelf.py:96

(self, dataset, num_replicas=None, rank=None, local_rank=None, local_size=None, shuffle=True)

lib/dataset/samplers.py:45

(self, dataset, num_replicas=None, rank=None, local_rank=None, local_size=None, shuffle=True)

lib/dataset/samplers.py:104

(self, cfg, image_set, is_train, transform=None)

lib/dataset/h36m.py:72

(self, cfg, image_set, is_train, transform=None)

lib/dataset/campus.py:73

(self, cfg, image_set, is_train, transform=None)

lib/dataset/JointsDataset.py:36

(self, cfg, image_set, is_train, transform=None)

lib/dataset/panoptic.py:220

lib/core/loss.py:61

(self, loss_type)

lib/core/loss.py:82

(self, loss_type)

lib/core/loss.py:120

(self, loss_type)

lib/core/loss.py:147

(self, loss_type)

lib/core/loss.py:198

(self, loss_type)

lib/core/loss.py:246

lib/core/function.py:591

size: <int> the number of nodes. root: <dict> the root NODE.

lib/structural/structural_triangulation.py:15

lib/mvn/utils/misc.py:15

(self, point, size=3, color=(0, 0, 255))

lib/mvn/utils/volumetric.py:9

(self, start_point, end_point, size=2, color=(0, 0, 255))

lib/mvn/utils/volumetric.py:26

(self, position, sides)

lib/mvn/utils/volumetric.py:45

(self, R, t, K, dist=None, name="")

lib/mvn/utils/multiview.py:6

h36m_root: Path to 'processed/' directory in Human3.6M labels_path: Path to 'human36m-multivi

lib/mvn/datasets/human36m.py:20

lib/mvn/models/loss.py:8

(self, threshold=400)

lib/mvn/models/loss.py:18

lib/mvn/models/loss.py:32

lib/mvn/models/loss.py:53

(self, inplanes, planes, stride=1, downsample=None)

lib/mvn/models/pose_resnet.py:60

(self, inplanes, planes, stride=1, downsample=None)

lib/mvn/models/pose_resnet.py:101

(self, in_channels, n_classes)

lib/mvn/models/pose_resnet.py:141

(self, block, layers, num_joints, num_input_channels=3, deconv_with_bias=Fal

lib/mvn/models/pose_resnet.py:185

(self, in_planes, out_planes, kernel_size)

lib/mvn/models/v2v.py:8

← previous next →201–300 of 561, ranked by callers