Functions — github.com/drinkingcoder/NeuralMarker

hub / github.com/drinkingcoder/NeuralMarker / functions

Functions183 in github.com/drinkingcoder/NeuralMarker

⨍ Functions 183◇ Types & classes 40

↓ 14 callersFunctioncoords_grid

(batch, ht, wd, device='cpu')

core/utils/utils.py:158

↓ 8 callersFunctionblend

(out, source, scene, blend_type, mask=None, use_colormap=False)

evaluation_DVL.py:59

↓ 8 callersFunctionimage_flow_warp

Input: image: HxWx3 numpy flow: HxWx2 torch.Tensor Output: outImg: HxWx3 numpy

evaluation_DVL.py:36

↓ 7 callersMethod__init__

(self, hidden_dim=128, input_dim=192+128)

core/update.py:27

↓ 7 callersMethod__init__

(self, output_size=(480, 640), geometric_model='affine', ra

core/utils/transformation.py:673

↓ 6 callersFunctionresize_lighting

(img)

eval_utils.py:22

↓ 5 callersFunctioneval

Args: @id: (int) marker image id, for logging @scene_images: (list) test scene images @marker: (array) marker image

evaluation_DVL.py:355

↓ 4 callersFunctionexpand_dim

(tensor, dim, desired_dim_len)

core/utils/utils.py:201

↓ 4 callersFunctionresize_viewpoint

(img)

eval_utils.py:15

↓ 4 callersFunctionrgb_to_srgb

(rgb)

harsh_lighting_utils.py:19

↓ 4 callersFunctionsample_one

Input: -img (N, C, H, W) (flow) -shiftx, shifty (N, c, H, W)

core/utils/utils.py:35

↓ 4 callersMethodsample_one

Input: -img (N, C, H, W) -shiftx, shifty (N, c, H, W)

core/utils/forward_warp.py:74

↓ 3 callersMethod__init__

(self, output_dim=128, norm_fn='batch', dropout=0.0)

core/extractor.py:119

↓ 3 callersMethod_make_layer

(self, dim, stride=1)

core/extractor.py:159

↓ 3 callersMethod_make_layer

(self, dim, stride=1)

core/extractor.py:235

↓ 3 callersMethodestimate

Input: im1, im2: HxWx3 numpy (it is not necessary for im1 and im2 to be same shape, because im2 will be resized to th

flow_estimator.py:23

↓ 3 callersMethodforward_single

(self, x, data=None, layer=2)

core/encoders.py:19

↓ 3 callersFunctionimage_flow_warp

Input: image: HxWx3 numpy flow: HxWx2 torch.Tensor Output: outImg: HxWx3 numpy

core/utils/utils.py:344

↓ 3 callersMethodsymmetricImagePad

(self, image_batch, padding_factor)

core/utils/transformation.py:206

↓ 3 callersFunctionupflow8

(flow, mode='bilinear')

core/utils/utils.py:164

↓ 2 callersFunctionPSNR

(image_true, image_test, mask, heatmap=False, with_mask=True)

metrics.py:17

↓ 2 callersFunctionSSIM

(im1, im2, mask, multichannel=True, heatmap=False, with_mask=True)

metrics.py:62

↓ 2 callersFunctionblend

(estimator, marker, scene, frame, args, warp = 'homography')

demo_video.py:72

↓ 2 callersFunctioncompute_symmetrical_epipolar_distance_loss

(args, flow_ab, Fm)

core/loss.py:39

↓ 2 callersFunctioncompute_transformation_loss

(args, data, flow_est, type='BB1')

core/loss.py:115

↓ 2 callersFunctioncoords_grid

(batch, ht, wd)

evaluation_DVL.py:32

↓ 2 callersFunctioncoords_grid

(batch, ht, wd)

demo_video.py:43

↓ 2 callersFunctionentry_convert

(entry)

core/utils/utils.py:169

↓ 2 callersFunctionget_colormap

(flow, H, W)

colormap.py:76

↓ 2 callersFunctionget_continuous_color

Plotly continuous colorscales assign colors to the range [0, 1]. This function computes the intermediate color for any value in that range.

colormap.py:20

↓ 2 callersFunctionhomography_mat_from_4_pts

(theta)

core/utils/transformation.py:490

↓ 2 callersFunctionimage_flow_warp

Input: image: HxWx3 numpy flow: HxWx2 torch.Tensor Output: outImg: HxWx3 numpy

demo_video.py:49

↓ 2 callersMethodinitialize_flow

Flow is represented as difference between two coordinate grids flow = coords1 - coords0

core/biraft.py:70

↓ 2 callersFunctionmetrics

(args, id, output_images, masks, scene_images=None, source=None)

metrics.py:142

↓ 2 callersFunctionreadFlow

Read .flo file in Middlebury format

core/utils/frame_utils.py:12

↓ 2 callersFunctionread_video

(video_path)

demo_video.py:20

↓ 2 callersFunctionsave_checkpoint

(args, model, optimizer=None, step=None, epoch=None)

train.py:42

↓ 2 callersMethodupsample_flow

Upsample flow field [H/8, W/8, 2] -> [H, W, 2] using convex combination

core/biraft.py:79

↓ 1 callersMethod__init__

(self, pretrained=True)

core/encoders.py:7

↓ 1 callersMethod__init__

(self, path, Tcw, K)

core/datasets.py:27

↓ 1 callersMethodapply_transformation

(self, theta, points)

core/utils/transformation.py:586

↓ 1 callersFunctionbilinear_sampler

Wrapper for grid_sample, uses pixel coordinates

core/utils/utils.py:141

↓ 1 callersFunctionblend_RANSAC

(scene_images, marker, coarseModel=None, network=None, source=None, blend_type='D', use_colormap=False)

evaluation_DVL.py:179

↓ 1 callersFunctionblend_SPSG

(scene_images, marker, matching=None, source=None, blend_type='D', use_colormap=True)

evaluation_DVL.py:301

↓ 1 callersFunctionblend_homography

(scene_images, marker, source=None, blend_type='D', detector='SIFT', use_colormap=True)

evaluation_DVL.py:238

↓ 1 callersFunctionblend_life

(scene_images, marker, estimator, source=None, blend_type='D', warp='grid_sample', use_colormap=True)

evaluation_DVL.py:112

↓ 1 callersFunctionblend_pdc

(scene_images, marker, flow_estimator, estimate_uncertainty, source=None, blend_type='D', use_colormap=True)

evaluation_DVL.py:73

↓ 1 callersFunctioncleanup

()

train.py:39

↓ 1 callersMethodcompute_L_inverse

(self, X, Y)

core/utils/transformation.py:566

↓ 1 callersFunctioncompute_all_loss

(args, data, flows_pre)

core/loss.py:8

↓ 1 callersFunctionconfig_prase

(args)

train.py:179

↓ 1 callersFunctionconvert_flow_to_mapping

(flow, output_channel_first=True)

core/utils/utils.py:244

↓ 1 callersMethodcorr

(fmap1, fmap2)

core/corr.py:53

↓ 1 callersFunctiondemo

()

demo_video.py:155

↓ 1 callersFunctionepipolar_distance

copy from kornia

core/loss.py:87

↓ 1 callersMethodexpandImagePad

(self, image_batch, padding_factor)

core/utils/transformation.py:224

↓ 1 callersFunctionfetch_dataloader

(args, split='train')

core/datasets.py:237

↓ 1 callersFunctionflow_uv_to_colors

Applies the flow color wheel to (possibly clipped) flow components u and v. According to the C++ source code of Daniel Scharstein Accord

core/utils/flow_viz.py:70

↓ 1 callersMethodforward

Estimate optical flow between pair of frames

core/raft.py:91

↓ 1 callersFunctionfundamental_matrix_gen

(Tcw1, Tcw2, K1, K2)

core/utils/utils.py:180

↓ 1 callersFunctionget_color

(colorscale_name, loc)

colormap.py:7

↓ 1 callersFunctionget_demo_video_args

()

config.py:73

↓ 1 callersFunctionget_eval_args

()

config.py:4

↓ 1 callersFunctionget_gaussian_weights

(x, y, x1, x2, y1, y2)

core/utils/utils.py:27

↓ 1 callersMethodget_gaussian_weights

(self, x, y, x1, x2, y1, y2)

core/utils/forward_warp.py:65

↓ 1 callersFunctionget_life_args

()

config.py:63

↓ 1 callersFunctionget_plotly_colors

(num_points, colorscale)

colormap.py:68

↓ 1 callersFunctionget_twins_args

()

config.py:50

↓ 1 callersMethodimage_pair_process

(self, img1, img2)

flow_estimator.py:52

↓ 1 callersMethodinference_model

(self, im1, im2)

flow_estimator.py:65

↓ 1 callersMethodinitialize_flow

Flow is represented as difference between two coordinate grids flow = coords1 - coords0

core/raft.py:68

↓ 1 callersFunctionload_checkpoint

(args, model, optimizer)

train.py:60

↓ 1 callersFunctionmake_colorwheel

Generates a color wheel for optical flow visualization as presented in: Baker et al. "A Database and Evaluation Methodology for Optical F

core/utils/flow_viz.py:20

↓ 1 callersMethodpad

(self, *inputs)

core/utils/utils.py:18

↓ 1 callersMethodrandom_select

(self)

core/datasets.py:195

↓ 1 callersFunctionreadPFM

(file)

core/utils/frame_utils.py:33

↓ 1 callersFunctionrefine_grid

(grid)

core/utils/utils.py:237

↓ 1 callersFunctionrender

(decompose_result, replace_result, poster_brightness: float, data_root: str, mask = None, save=False)

harsh_lighting_utils.py:30

↓ 1 callersFunctionrun

()

evaluation_DVL.py:462

↓ 1 callersFunctionsave_video

(imgs, size, video_path, fps=30)

demo_video.py:35

↓ 1 callersFunctionsetup

(args, rank, world_size)

train.py:32

↓ 1 callersFunctionsymmetrical_epipolar_distance

copy from kornia

core/loss.py:57

↓ 1 callersFunctionsynthesis_data

(args)

synthesis_datasets.py:15

↓ 1 callersMethodunpad

(self,x)

core/utils/utils.py:21

↓ 1 callersMethodupsample_flow

Upsample flow field [H/8, W/8, 2] -> [H, W, 2] using convex combination

core/raft.py:77

↓ 1 callersFunctionvalidate

(args, model, step=None, split=None)

evaluation_FM.py:16

↓ 1 callersFunctionwriteFlow

Write optical flow to file. If v is None, uv is assumed to contain both u and v channels, stacked in depth. Original code by Deqing

core/utils/frame_utils.py:70

Input: im: Torch.tensor[C, H, W] tnf_type: 'random', 'affine', 'hom', 'tps' Output: im: Torch.tensor[C, H, W]

core/utils/augmentor.py:15

Method__call__

(self, image_batch, theta_aff, theta_aff_tps, use_cuda=True)

core/utils/transformation.py:36

Method__call__

(self, image_batch, theta_batch=None, out_h=None, out_w=None, return_warped_image=True, retur

core/utils/transformation.py:105

Method__call__

(self, batch)

core/utils/transformation.py:167

Method__call__

(self, batch)

core/utils/transformation.py:240

Method__call__

(self, batch)

core/utils/transformation.py:286

Method__call__

(self, batch)

core/utils/transformation.py:344

Method__call__

(self)

core/utils/transformation.py:692

next →1–100 of 183, ranked by callers