Functions — github.com/NVlabs/CenterPose

hub / github.com/NVlabs/CenterPose / functions

Functions615 in github.com/NVlabs/CenterPose

⨍ Functions 615◇ Types & classes 95

↓ 234 callersMethodappend

(self, hit_miss, num_instances)

src/tools/objectron_eval/objectron/dataset/metrics.py:44

↓ 206 callersMethodwrite

(self, txt)

src/lib/logger.py:63

↓ 36 callersMethodappend

(self, hit_miss, num_instances)

data/objectron/dataset/metrics.py:44

↓ 31 callersMethodupdate

(self, val, n=1)

src/lib/utils/utils.py:18

↓ 28 callersMethodcompute_ap_curve

Computes the precision/recall curve.

src/tools/objectron_eval/objectron/dataset/metrics.py:86

↓ 25 callersMethodrecord_hit_miss

Records the hit or miss for the object based on the metric threshold.

src/tools/objectron_eval/objectron/dataset/metrics.py:19

↓ 20 callersFunction_transpose_and_gather_feat

(feat, ind)

src/lib/models/utils.py:43

↓ 17 callersFunctiondraw_umich_gaussian

(heatmap, center, radius, k=1)

src/lib/utils/image.py:135

↓ 16 callersMethodappend

(self, hit_miss, num_instances)

src/tools/objectron_eval/objectron/dataset/metrics_nvidia.py:59

↓ 16 callersMethodrun

(self, image_or_path_or_tensor, filename=None, meta_inp={}, preprocessed_flag=False)

src/lib/detectors/base_detector.py:390

↓ 15 callersFunctionaffine_transform

(pt, t)

src/lib/utils/image.py:71

↓ 12 callersMethodadd_coco_hp

(self, points, img_id='default', pred_flag='pred', PAPER_DISPLAY=False)

src/lib/utils/debugger.py:214

↓ 11 callersMethod__init__

(self, levels, channels, num_classes=1000, block=BasicBlock, residual_root=False, linear_root

src/lib/models/networks/pose_dla_dcn.py:228

↓ 11 callersMethodadd_blend_img

(self, back, fore, img_id='blend', trans=0.6)

src/lib/utils/debugger.py:63

↓ 10 callersMethodadd_arrow

(self, st, ed, img_id, c=(255, 0, 255), w=2)

src/lib/utils/debugger.py:324

↓ 10 callersMethodload_pretrained_model

(self, data='imagenet', name='dla34', hash='ba72cf86')

src/lib/models/networks/dlav0.py:333

↓ 9 callersMethod__init__

(self, levels, channels, num_classes=1000, block=BasicBlock, residual_root=False, return_leve

src/lib/models/networks/dlav0.py:223

↓ 9 callersMethodbackward

(ctx, grad_output)

src/lib/models/networks/DCNv2/dcn_v2.py:37

↓ 8 callersMethod__init__

(self)

src/lib/models/losses.py:259

↓ 8 callersMethodadd_coco_bbox

(self, bbox, cat, conf=1, id=None, show_txt=True, img_id='default')

src/lib/utils/debugger.py:131

↓ 8 callersMethodadd_obj_scale

(self, bbox, scale, img_id='default', pred_flag='pred')

src/lib/utils/debugger.py:165

↓ 8 callersFunctionget_affine_transform

(center, scale, rot, output_size,

src/lib/utils/image.py:35

↓ 8 callersFunctionrun

(c, command)

src/tools/objectron_eval/shell_eval_image_CenterPose.py:21

↓ 8 callersFunctionsave_model

(path, epoch, model, optimizer=None)

src/lib/models/model.py:90

↓ 8 callersFunctiontest_main

(c, mode)

src/tools/objectron_eval/shell_eval_video_CenterPose.py:48

↓ 8 callersFunctiontest_main

(c, mode)

src/tools/objectron_eval/shell_eval_video_CenterPoseTrack.py:60

↓ 7 callersMethodadd_img

(self, img, img_id='default', revert_color=False)

src/lib/utils/debugger.py:58

↓ 7 callersMethodgather

(self, outputs, output_device)

src/lib/models/data_parallel.py:83

↓ 6 callersMethodGetIntersection

(self, fDst1, fDst2, P1, P2)

data/bbox_collision_detection.py:32

↓ 6 callersMethodInBox

(self, Axis)

data/bbox_collision_detection.py:45

↓ 6 callersMethod_trans_bbox

Transform bounding boxes according to image crop.

src/lib/detectors/base_detector.py:80

↓ 6 callersMethodgen_colormap

(self, img, output_res=None, color=None)

src/lib/utils/debugger.py:75

↓ 6 callersMethodpredict

Predict the box's 2D and 3D keypoint from the input images. Note that the predicted 3D bounding boxes are correct up to an scale.

src/tools/objectron_eval/eval_video_official.py:317

↓ 6 callersMethodreset_tracking

(self)

src/lib/detectors/base_detector.py:774

↓ 6 callersFunctionsafe_divide

(i1, i2)

src/tools/objectron_eval/group_report_new.py:43

↓ 5 callersMethod__init__

( self, n, nstack, dims, modules, heads, pre=None, cnv_dim=256, make_tl_layer=None, ma

src/lib/models/networks/large_hourglass.py:191

↓ 5 callersFunctioncreate_model

(arch, heads, head_conv, opt=None)

src/lib/models/model.py:26

↓ 5 callersMethodfit

Estimates a box 9-dof parameters from the given vertices. Directly computes the scale of the box, then solves for orientation and tra

src/tools/objectron_eval/objectron/dataset/box.py:130

↓ 5 callersFunctiongaussian_radius

(det_size, min_overlap=0.7)

src/lib/utils/image.py:103

↓ 5 callersMethodgen_colormap_hp

(self, img, output_res=None)

src/lib/utils/debugger.py:116

↓ 5 callersMethodimg_summary

(self, tag, value_list, step)

src/lib/logger.py:82

↓ 5 callersMethodinit

(self, opt)

src/lib/opts.py:431

↓ 5 callersFunctionload_model

(model, model_path, optimizer=None, resume=False, lr=None, lr_step=None)

src/lib/models/model.py:34

↓ 5 callersMethodparse

(self, opt)

src/lib/opts.py:330

↓ 5 callersFunctionpnp_shell

(opt, meta, bbox, points_filtered, scale, OPENCV_RETURN = False)

src/lib/utils/pnp/cuboid_pnp_shell.py:11

↓ 5 callersFunctionsafe_divide

(i1, i2)

src/tools/objectron_eval/eval_video_official.py:105

↓ 5 callersFunctionsafe_divide

(i1, i2)

src/tools/objectron_eval/eval_image_official.py:107

↓ 5 callersFunctionsafe_divide

(i1, i2)

src/tools/objectron_eval/group_report_old.py:40

↓ 5 callersMethodscalar_summary

Log a scalar variable.

src/lib/logger.py:77

↓ 5 callersMethodscatter

(self, inputs, kwargs, device_ids, chunk_sizes)

src/lib/models/data_parallel.py:77

↓ 5 callersMethodval

(self, epoch, data_loader)

src/lib/trains/base_trainer.py:154

↓ 4 callersFunction_gather_feat

(feat, ind, mask=None)

src/lib/models/utils.py:14

↓ 4 callersMethod_make_layer

(self, block, planes, blocks, stride=1)

src/lib/models/networks/msra_resnet.py:178

↓ 4 callersMethod_make_layer

(self, block, planes, blocks, stride=1)

src/lib/models/networks/resnet_dcn.py:183

↓ 4 callersFunction_nms

(heat, kernel=3)

src/lib/models/decode.py:17

↓ 4 callersMethodadd_axes

(self, box, cam_intrinsic, img_id='default')

src/lib/utils/debugger.py:299

↓ 4 callersMethodcheck

(self)

data/bbox_collision_detection.py:54

↓ 4 callersMethodclose

(self)

src/lib/logger.py:74

↓ 4 callersMethodcompute_ap_curve

Computes the precision/recall curve.

data/objectron/dataset/metrics.py:86

↓ 4 callersFunctioncompute_res_loss

(output, target)

src/lib/models/losses.py:279

↓ 4 callersMethodfit

Estimates a box 9-dof parameters from the given vertices. Directly computes the scale of the box, then solves for orientation and translation

data/objectron/dataset/box.py:126

↓ 4 callersMethodparse_example

Parses image and label from a tf.Example proto. Args: example: A tf.Example proto. Returns: A tuple of image and

src/tools/objectron_eval/objectron/dataset/parser.py:93

↓ 4 callersMethodpredict

Implement your own function/model to predict the box's 2D and 3D keypoint from the input images. Note that the predicte

src/tools/objectron_eval/eval_image_official.py:198

↓ 4 callersMethodrecord_hit_miss

Records the hit or miss for the object based on the metric threshold.

data/objectron/dataset/metrics.py:19

↓ 4 callersFunctionsafe_divide

(i1, i2)

data/objectron/dataset/eval.py:43

↓ 4 callersFunctionsafe_divide

(i1, i2)

src/tools/objectron_eval/objectron/dataset/eval.py:44

↓ 4 callersMethodset_device

(self, gpus, chunk_sizes, device)

src/lib/trains/base_trainer.py:42

↓ 4 callersFunctionswapPositions

(list, pos1, pos2)

data/utils.py:56

↓ 4 callersFunctiontransform_preds

(coords, center, scale, output_size)

src/lib/utils/image.py:23

↓ 4 callersMethodupdate_dataset_info_and_set_heads

(self, opt, dataset)

src/lib/opts.py:378

↓ 3 callersMethod__init__

(self, in_channels, out_channels, kernel_size, stride, padding, dilation=1,

src/lib/models/networks/DCNv2/dcn_v2.py:99

↓ 3 callersMethoddebug

(self, batch, output, iter_id)

src/lib/trains/base_trainer.py:145

↓ 3 callersMethodinit_track

(self, meta)

src/lib/utils/tracker.py:21

↓ 3 callersMethodinside

Tests whether a given point is inside the box. Brings the 3D point into the local coordinate of the box. In the local coordinate, the loo

data/objectron/dataset/box.py:161

↓ 3 callersMethodinside

Tests whether a given point is inside the box. Brings the 3D point into the local coordinate of the box. In the local coordinate,

src/tools/objectron_eval/objectron/dataset/box.py:165

↓ 3 callersMethodiou

Computes the exact IoU using Sutherland-Hodgman algorithm.

src/tools/objectron_eval/objectron/dataset/iou.py:22

↓ 3 callersMethodparse_camera

Parses camera from a tensorflow example.

src/tools/objectron_eval/objectron/dataset/parser.py:195

↓ 3 callersMethodparse_plane

Parses plane from a tensorflow example.

src/tools/objectron_eval/objectron/dataset/parser.py:218

↓ 3 callersFunctionpreprocess

(annotation_file,category,opt)

data/preprocess.py:83

↓ 3 callersMethodreset

(self)

src/lib/utils/tracker.py:51

↓ 3 callersMethodsample

Samples a 3D point uniformly inside this box.

src/tools/objectron_eval/objectron/dataset/box.py:185

↓ 3 callersMethodscaled_axis_aligned_vertices

Returns an axis-aligned set of verticies for a box of the given scale. Args: scale: A 3*1 vector, specifiying the size of the box in x-y-z

data/objectron/dataset/box.py:109

↓ 3 callersMethodscaled_axis_aligned_vertices

Returns an axis-aligned set of verticies for a box of the given scale. Args: scale: A 3*1 vector, specifiying the size of the box i

src/tools/objectron_eval/objectron/dataset/box.py:113

↓ 3 callersMethodstep

(self, dets, boxes=[])

src/lib/utils/tracker.py:112

↓ 3 callersMethodtrain

(self, epoch, data_loader)

src/lib/trains/base_trainer.py:157

↓ 2 callersFunctionDataParallel

(module, device_ids=None, output_device=None, dim=0, chunk_sizes=None)

src/lib/models/data_parallel.py:120

↓ 2 callersFunctionDataloader

(opt)

src/tools/objectron_eval/eval_video_official.py:131

↓ 2 callersMethod__init__

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

src/lib/models/networks/msra_resnet.py:38

↓ 2 callersMethod__init__

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

src/lib/models/networks/resnet_dcn.py:42

↓ 2 callersMethod_classify_point_to_plane

Classify position of a point w.r.t the given plane. See Real-Time Collision Detection, by Christer Ericson, page 364. Args: point: 3x1

data/objectron/dataset/iou.py:188

↓ 2 callersMethod_classify_point_to_plane

Classify position of a point w.r.t the given plane. See Real-Time Collision Detection, by Christer Ericson, page 364. Args:

src/tools/objectron_eval/objectron/dataset/iou.py:191

↓ 2 callersMethod_clip_poly

Clips the polygon with the plane using the Sutherland-Hodgman algorithm. See en.wikipedia.org/wiki/Sutherland-Hodgman_algorithm for the overview

data/objectron/dataset/iou.py:100

↓ 2 callersMethod_clip_poly

Clips the polygon with the plane using the Sutherland-Hodgman algorithm. See en.wikipedia.org/wiki/Sutherland-Hodgman_algorithm for the overv

src/tools/objectron_eval/objectron/dataset/iou.py:103

↓ 2 callersMethod_compute_intersection_points

Computes the intersection of two boxes.

data/objectron/dataset/iou.py:72

↓ 2 callersMethod_compute_intersection_points

Computes the intersection of two boxes.

src/tools/objectron_eval/objectron/dataset/iou.py:75

↓ 2 callersMethod_get_aug_param

(self, c_ori, s, width, height, disturb=False)

src/lib/datasets/dataset_combined.py:240

↓ 2 callersMethod_get_border

(self, border, size)

src/lib/datasets/dataset_combined.py:233

↓ 2 callersMethod_get_input

(self, img, trans_input)

src/lib/datasets/dataset_combined.py:278

↓ 2 callersMethod_get_rotated_box

Rotate a box along its vertical axis. Args: box: Input box. angle: Rotation angle in rad. Returns: A rot

src/tools/objectron_eval/eval_video_official.py:1066

↓ 2 callersMethod_intersect

Computes the intersection of a line with an axis-aligned plane. Args: plane: Formulated as two 3D points on the plane. prev_point: Th

data/objectron/dataset/iou.py:159

next →1–100 of 615, ranked by callers