Functions — github.com/amjltc295/Free-Form-Video-Inpainting

hub / github.com/amjltc295/Free-Form-Video-Inpainting / functions

Functions471 in github.com/amjltc295/Free-Form-Video-Inpainting

⨍ Functions 471◇ Types & classes 106

↓ 109 callersMethodappend

(self, file_)

src/utils/readers.py:116

↓ 20 callersMethodsave

(self,file='index')

libs/PerceptualSimilarity/util/html.py:47

↓ 15 callersFunctionget_everything_under

(root_dir, pattern='*', only_dirs=False, only_files=False)

src/utils/util.py:235

↓ 12 callersMethod__init__

( self, in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, bia

src/model/blocks.py:257

↓ 11 callersMethod__init__

(self)

src/model/loss.py:292

↓ 11 callersFunctionmake_dirs

(dir_name)

src/utils/util.py:38

↓ 8 callersMethod__init__

(self, dir_name, read=True, max_length=None, sample_period=1)

src/utils/readers.py:93

↓ 8 callersMethodforward

(self, in0, in1)

libs/PerceptualSimilarity/models/networks_basic.py:229

↓ 8 callersFunctionget_instance

(module, name, config, *args)

src/train.py:19

↓ 8 callersMethodinitialize

INPUTS model - ['net-lin'] for linearly calibrated network ['net'] for off-the-shelf network

libs/PerceptualSimilarity/models/dist_model.py:24

↓ 6 callersFunctionread_dirnames_under_root

(root_dir, skip_list=[])

src/utils/util.py:59

↓ 5 callersMethod__init__

(self, pnet_type='vgg', pnet_rand=False, use_gpu=True)

libs/PerceptualSimilarity/models/networks_basic.py:18

↓ 5 callersMethodbackward

(ctx, grad_output)

src/model/tsm_utils.py:24

↓ 5 callersMethodforward

Pred and target are Variables. If normalize is on, assumes the images are between [0,1] and then scales thembetween [-1, 1] I

libs/PerceptualSimilarity/__init__.py:16

↓ 4 callersMethod__init__

(self, nc_in, nc_out, nf, use_bias, norm, conv_by, conv_type, use_skip_connection: bool = False)

src/model/networks.py:14

↓ 4 callersMethod_get_gan_loss

(self, outputs, target, masks, discriminator, w, guidances=None, is_disc=None)

src/trainer/trainer.py:212

↓ 4 callersMethodadd_images

(self, ims, txts, links, width=400)

libs/PerceptualSimilarity/util/html.py:35

↓ 4 callersMethodforward_pair

(self,in1,in2,retPerLayer=False)

libs/PerceptualSimilarity/models/dist_model.py:101

↓ 4 callersFunctionget_every

(d)

src/scripts/mask_scripts/apply_mask.py:30

↓ 4 callersFunctionget_i3d_activations

Get features from i3d model and flatten them to 1d feature, valid target endpoints are defined in InceptionI3d.VALID_ENDPOINTS VALID_END

src/evaluate.py:306

↓ 4 callersMethodwrite_files_to_video

(self, output_filename, fps=DEFAULT_FPS, frame_num_when_repeat_list=[1])

src/utils/readers.py:190

↓ 4 callersFunctionzoom_to_res

(img,res=256,order=0,axis=0)

libs/PerceptualSimilarity/util/visualizer.py:11

↓ 3 callersMethod__init__

( self, rids: RootInputDirectories, rods: RootOutputDirectories, args: dict,

src/data_loader/dataset.py:26

↓ 3 callersMethod__init__

(self, conv_type)

src/model/modules.py:16

↓ 3 callersMethod__init__

(self, requires_grad=False, pretrained=True)

libs/PerceptualSimilarity/models/pretrained_networks.py:97

↓ 3 callersMethod_process_data

(self, data)

src/trainer/trainer.py:260

↓ 3 callersMethod_unpack_data

(self, data_input, model_output)

src/trainer/trainer.py:276

↓ 3 callersMethod_write_images

( self, inputs, outputs, targets, masks, output_edges=None, target_edges=None, model_o

src/trainer/trainer.py:165

↓ 3 callersMethodcompute_pad

(self, dim, s)

src/model/i3d.py:10

↓ 3 callersMethodcompute_pad

(self, dim, s)

src/model/i3d.py:69

↓ 3 callersMethodforward

(self)

libs/PerceptualSimilarity/perceptual_loss.py:37

↓ 3 callersMethodforward

Function computes the distance between image patches in0 and in1 INPUTS in0, in1 - torch.Tensor object of shape Nx3xXxY - image p

libs/PerceptualSimilarity/models/dist_model.py:107

↓ 3 callersFunctionget_edge

(tensor)

src/utils/edge.py:15

↓ 3 callersMethodload_data

()

libs/PerceptualSimilarity/data/base_data_loader.py:9

↓ 3 callersFunctionmake_dir_under_root

(root_dir, name)

src/utils/util.py:47

↓ 3 callersMethodpostprocess

(self, masked_imgs, masks, c11)

src/model/networks.py:39

↓ 3 callersMethodsave_files

(self, output_dir=None)

src/utils/readers.py:131

↓ 3 callersMethodset_step

(self, step, mode='train')

src/utils/visualization.py:28

↓ 3 callersMethodshow

(self)

src/utils/screen.py:16

↓ 2 callersFunctionLaguerre_Gauss_Circular_Harmonic_1_0

(size, sigma)

libs/video_quality/reco.py:46

↓ 2 callersFunctionLaguerre_Gauss_Circular_Harmonic_3_0

(size, sigma)

libs/video_quality/reco.py:36

↓ 2 callersMethod__init__

Initializes Unit3D module.

src/model/i3d.py:37

↓ 2 callersMethod__init__

(self, df=32)

src/model/completion_model.py:82

↓ 2 callersMethod_cut_to_align

Cut the boarder by 1 frame(temporal)/row(height)/col(width) such that dimensions of x and y could be the same.

src/model/completion_model.py:36

↓ 2 callersFunction_decide_patch_num

(box_width, patch_size)

src/utils/util.py:112

↓ 2 callersMethod_eval_metrics

(self, output, target)

src/trainer/trainer.py:205

↓ 2 callersMethod_get_mask_name

(self, index)

src/data_loader/dataset.py:157

↓ 2 callersMethod_get_non_gan_loss

(self, data_input, model_output)

src/trainer/trainer.py:286

↓ 2 callersMethodadd_header

(self, str)

libs/PerceptualSimilarity/util/html.py:27

↓ 2 callersFunctionblock_view

Provide a 2D block view to 2D array. No error checking made. Therefore meaningful (as implemented) only for blocks strictly compatible with th

libs/video_quality/ssim.py:27

↓ 2 callersFunctioncluster_by_masked_area

(root_dir, args)

src/scripts/mask_scripts/gen_masks.py:182

↓ 2 callersMethodconcat_feature

(self, ca, cb)

src/model/modules.py:123

↓ 2 callersFunctiondecide_nStroke

(args)

src/scripts/mask_scripts/gen_masks.py:217

↓ 2 callersFunctiondraw_mask_by_control_points

(mask, Xs, Ys, brushWidth, fill=255)

src/utils/mask_generators.py:178

↓ 2 callersFunctioneco

(img)

libs/video_quality/reco.py:73

↓ 2 callersFunctionerase_mask_boarder

(mask, gap)

src/scripts/mask_scripts/gen_masks.py:153

↓ 2 callersFunctionestimate_aggd_params

(x)

libs/video_quality/niqe.py:89

↓ 2 callersFunctionevaluate_fid_score

(root_gt_dir, root_result_dir, result_postfix)

src/evaluate.py:249

↓ 2 callersMethodevaluate_test_set

(self, output_root_dir=None, epoch=None, log=None)

src/trainer/trainer.py:411

↓ 2 callersFunctionevaluate_video_error

( result_images, gt_images, masks, flownet_checkpoint_path: str, evaluate_warping_error=True, prin

src/evaluate.py:143

↓ 2 callersMethodfill_fix_offset

(more_fix_crop, image_w, image_h, crop_w, crop_h)

src/data_loader/transform.py:184

↓ 2 callersMethodforward

(self, X)

src/model/vgg.py:28

↓ 2 callersMethodget_bboxeses

(self)

src/utils/readers.py:245

↓ 2 callersFunctionget_every

(d)

src/scripts/reorder.py:25

↓ 2 callersFunctionget_every

(d)

src/scripts/leave_32frames.py:26

↓ 2 callersFunctionget_fid_score

Given two distribution of features, compute the FID score between them

src/evaluate.py:341

↓ 2 callersMethodget_image_dir

(self)

libs/PerceptualSimilarity/util/html.py:24

↓ 2 callersFunctionget_landmarks_contour

(image, landmarks_predictor_path=DEFAULT_PATH)

src/utils/face.py:87

↓ 2 callersFunctionget_loader

(type, ratio)

src/scripts/dataset_json_generator.py:9

↓ 2 callersFunctionget_masked_ratio

Calculate the masked ratio. mask: Expected a binary PIL image, where 0 and 1 represent masked(invalid) and valid pixel values.

src/utils/mask_generators.py:210

↓ 2 callersFunctionget_video_masks_by_moving_random_stroke

Get video masks by random strokes which move randomly between each frame, including the whole stroke and its control points Parameters

src/utils/mask_generators.py:6

↓ 2 callersMethodgram_matrix

(self, input)

src/model/loss.py:69

↓ 2 callersFunctionim2tensor

(image, imtype=np.uint8, cent=1., factor=255./2.)

libs/PerceptualSimilarity/util/util.py:144

↓ 2 callersFunctionimg_read_yuv

(src_file, width, height)

libs/video_quality/measure.py:34

↓ 2 callersFunctioninit_i3d_model

()

src/evaluate.py:35

↓ 2 callersMethodinitialize

(self, datafolders, dataroot='./dataset',dataset_mode='2afc',load_size=64,batch_size=1,serial_batches=True, nT

libs/PerceptualSimilarity/data/custom_dataset_data_loader.py:23

↓ 2 callersFunctionmkdir

(path)

libs/PerceptualSimilarity/util/util.py:252

↓ 2 callersMethodname

(self)

libs/PerceptualSimilarity/models/dist_model.py:21

↓ 2 callersFunctionnormalize_blob

(in_feat,eps=1e-10)

libs/PerceptualSimilarity/util/util.py:63

↓ 2 callersFunctionnormalize_tensor

(in_feat,eps=1e-10)

libs/PerceptualSimilarity/util/util.py:74

↓ 2 callersFunctionnp2tensor

(np_obj)

libs/PerceptualSimilarity/util/util.py:95

↓ 2 callersMethodpreprocess

(self, masked_imgs, masks, guidances)

src/model/networks.py:22

↓ 2 callersFunctionrandom_accelerate

(velocity, maxAcceleration, dist='uniform')

src/utils/mask_generators.py:91

↓ 2 callersFunctionread_filenames_from_dir

(dir_name, reader, max_length=None)

src/utils/util.py:247

↓ 2 callersFunctionrm_dirs

(dir_name, ignore_errors=False)

src/utils/util.py:53

↓ 2 callersMethodsave

(self, label)

libs/PerceptualSimilarity/models/base_model.py:31

↓ 2 callersFunctionsave_frames_to_dir

(frames, dirname)

src/utils/readers.py:502

↓ 2 callersMethodsave_network

(self, network, path, network_label, epoch_label)

libs/PerceptualSimilarity/models/base_model.py:35

↓ 2 callersMethodset_files

(self, files)

src/utils/readers.py:119

↓ 2 callersMethodtrain

Full training logic

src/base/base_trainer.py:93

↓ 2 callersFunctiontransform

(k, v)

src/scripts/FVI_checkpoint_compatibility_transform.py:58

↓ 2 callersFunctionwalk_clean

(root)

src/scripts/clean_saved.py:52

↓ 2 callersFunctionwarp_optical_flow

Modified from https://github.com/NVlabs/PWC-Net/blob/fc6ebf9a70a7387164df09a3a2070ba16f9c1ede/PyTorch/models/PWCNet.py # NOQA warp an im2 ba

src/utils/flow_utils.py:6

↓ 1 callersFunctionCreateDataset

(dataroots,dataset_mode='2afc',load_size=64,)

libs/PerceptualSimilarity/data/custom_dataset_data_loader.py:5

↓ 1 callersMethod_evaluate_data_loader

(self, epoch=None, output_root_dir=None, data_loader=None, name='test')

src/trainer/trainer.py:80

↓ 1 callersMethod_evaluate_test_video

(self, output, gt_frames, masks)

src/trainer/trainer.py:194

↓ 1 callersMethod_get_loss

(self, outputs, warped_outputs, non_occlusion_masks, masks)

src/model/loss.py:235

↓ 1 callersMethod_get_masks

(self, size, start_idx, end_idx, fg_dir)

src/data_loader/dataset.py:80

↓ 1 callersMethod_get_non_occlusion_mask

(self, targets, warped_targets)

src/model/loss.py:227

↓ 1 callersMethod_get_sample_index_from_video

(self, length)

src/data_loader/dataset.py:71

next →1–100 of 471, ranked by callers