MCPcopy Create free account

hub / github.com/Lizhuoling/UniMODE / functions

Functions454 in github.com/Lizhuoling/UniMODE

⨍ Functions 454◇ Types & classes 88
↓ 1 callersFunctioncompute_virtual_scale_from_focal_spaces
Computes the scaling factor of depth from f0, H0 to f, H Args: f (float): the desired [virtual] focal length (px) H (float):
model/util/math_util.py:582
↓ 1 callersFunctionconvert_3d_box_to_2d
Converts a 3D box to a 2D box via projection. Args: K (np.array): intrinsics matrix 3x3 bbox3d (flexible): [[X Y Z W H L]]
model/util/math_util.py:499
↓ 1 callersFunctionconvnext_base
(pretrained=True, in_22k=True, **kwargs)
model/modeling/backbone/convnext.py:186
↓ 1 callersMethodcreate_frustum
(self, img_shape_after_aug, feat_shape)
model/modeling/detector3d/detector3d.py:620
↓ 1 callersFunctiondefault_argument_parser
()
model/util/default.py:7
↓ 1 callersFunctiondla102
(pretrained=False, tricks=False, **kwargs)
model/modeling/backbone/dla.py:379
↓ 1 callersFunctiondla102x
(pretrained=False, **kwargs)
model/modeling/backbone/dla.py:392
↓ 1 callersFunctiondla102x2
(pretrained=False, **kwargs)
model/modeling/backbone/dla.py:401
↓ 1 callersFunctiondla169
(pretrained=False, **kwargs)
model/modeling/backbone/dla.py:410
↓ 1 callersFunctiondla34
(pretrained=False, tricks=False, **kwargs)
model/modeling/backbone/dla.py:313
↓ 1 callersFunctiondla46_c
(pretrained=False, **kwargs)
model/modeling/backbone/dla.py:325
↓ 1 callersFunctiondla46x_c
(pretrained=False, **kwargs)
model/modeling/backbone/dla.py:335
↓ 1 callersFunctiondla60
(pretrained=False, tricks=False, **kwargs)
model/modeling/backbone/dla.py:355
↓ 1 callersFunctiondla60x
(pretrained=False, **kwargs)
model/modeling/backbone/dla.py:369
↓ 1 callersFunctiondla60x_c
(pretrained=False, **kwargs)
model/modeling/backbone/dla.py:345
↓ 1 callersFunctiondo_train
(cfg, model, dataset_id_to_unknown_cats, dataset_id_to_src, resume=False)
tools/train_net.py:122
↓ 1 callersFunctiondraw_2d_box
(im, box, color=(0, 200, 200), thickness=1)
model/vis/vis.py:710
↓ 1 callersFunctiondraw_3d_box
(im, K, category, box3d, R, color=(0, 200, 200), thickness=1, draw_face=False, category_color=False, category_
model/vis/vis.py:646
↓ 1 callersFunctiondraw_transparent_polygon
(im, verts, blend=0.5, color=(0, 255, 255))
model/vis/vis.py:564
↓ 1 callersFunctiondraw_umich_gaussian
(heatmap, center, radius, k=1, ignore=False)
model/data/dataset_mapper.py:366
↓ 1 callersFunctiondw_conv3x3
3x3 convolution with padding
model/modeling/backbone/vovnet.py:102
↓ 1 callersMethodevaluateImg
Perform evaluation for single category and image Returns: dict (single image results)
model/evaluation/omni3d_evaluation.py:1504
↓ 1 callersMethodextract_cam_feat
(self, imgs, batched_inputs)
model/modeling/detector3d/detector3d.py:113
↓ 1 callersMethodextract_point_feat
(self, points)
model/modeling/detector3d/detector3d.py:132
↓ 1 callersMethodfix_init_weight
(self)
model/modeling/backbone/eva.py:626
↓ 1 callersMethodforward_features
(self, x)
model/modeling/backbone/convnext.py:132
↓ 1 callersMethodforward_features
(self, x, return_patch_tokens=False)
model/modeling/backbone/eva.py:660
↓ 1 callersMethodforward_ffn
(self, src)
model/modeling/detector3d/deformable_detr.py:182
↓ 1 callersMethodforward_ffn
(self, tgt)
model/modeling/detector3d/deformable_detr.py:297
↓ 1 callersMethodforward_post
(self, src, src_mask: Optional[Tensor] = None,
model/modeling/detector3d/detr_transformer.py:158
↓ 1 callersMethodforward_post
(self, tgt, memory, tgt_mask: Optional[Tensor] = None, memory_mask:
model/modeling/detector3d/detr_transformer.py:246
↓ 1 callersMethodforward_pre
(self, src, src_mask: Optional[Tensor] = None, src_key_padding_mask: O
model/modeling/detector3d/detr_transformer.py:173
↓ 1 callersMethodforward_pre
(self, tgt, memory, tgt_mask: Optional[Tensor] = None, memory_mask: Op
model/modeling/detector3d/detr_transformer.py:270
↓ 1 callersFunctiongaussian2D
(shape, sigma=1)
model/data/dataset_mapper.py:356
↓ 1 callersFunctiongaussian_radius
(height, width, min_overlap=0.7)
model/data/dataset_mapper.py:335
↓ 1 callersMethodgenerate_mask
(self, images)
model/modeling/detector3d/detector3d.py:226
↓ 1 callersFunctionget_basic_renderer
(cameras, width, height, use_color=False)
model/util/math_util.py:815
↓ 1 callersFunctionget_camera
(K, width, height, switch_hands=True, R=None, T=None)
model/util/math_util.py:785
↓ 1 callersFunctionget_cfg_defaults
(cfg)
model/config/config.py:6
↓ 1 callersFunctionget_extensions
()
model/deformable_ops/setup.py:23
↓ 1 callersMethodget_geometry
(self, Ks, B, img_shape_after_aug)
model/modeling/detector3d/detector3d.py:632
↓ 1 callersFunctionget_geometry_data
(geometry_filename)
data_preparation/objectron_prepare.py:84
↓ 1 callersMethodget_matching
Description: One-to-one matching for a single batch.
model/modeling/detector3d/detector3d.py:929
↓ 1 callersFunctionget_model_url
(data='imagenet', name='dla34', hash='ba72cf86')
model/modeling/backbone/dla.py:31
↓ 1 callersFunctionget_polygon_grid
(im, poly_verts)
model/vis/vis.py:542
↓ 1 callersMethodget_reference_points
(spatial_shapes, valid_ratios, device)
model/modeling/detector3d/deformable_detr.py:208
↓ 1 callersMethodget_transform
(self, image)
model/data/dataset_mapper.py:201
↓ 1 callersMethodget_valid_ratio
(self, mask)
model/modeling/detector3d/deformable_detr.py:92
↓ 1 callersFunctionhypersim_distance_to_depth
(npyDistance)
data_preparation/hypersim_prepare.py:116
↓ 1 callersMethodinference
(self, detector_out, batched_inputs, ori_img_resolution)
model/modeling/detector3d/detector3d.py:206
↓ 1 callersFunctioninference_on_dataset
Run model on the data_loader. Also benchmark the inference speed of `model.__call__` accurately. The model will be used in eval mode.
model/evaluation/omni3d_evaluation.py:527
↓ 1 callersMethodinit_weights
(self)
model/modeling/detector3d/center_head.py:68
↓ 1 callersMethodinit_weights
(self)
model/modeling/detector3d/depthnet.py:212
↓ 1 callersFunctioninterp_color
(dist, bounds=[0, 1], color_lo=(0,0, 250), color_hi=(0, 250, 250))
model/vis/vis.py:18
↓ 1 callersFunctioniou
Computes the amount of Intersection over Union (IoU) between two different sets of boxes. Args: box_a (array or tensor): Mx4 boxes, d
model/util/math_util.py:851
↓ 1 callersFunctionload_omni3d_json
(json_file, image_root, dataset_name, filter_settings, filter_empty=False)
model/data/datasets.py:350
↓ 1 callersMethodloss
(self, detector_out, batched_inputs, ori_img_resolution)
model/modeling/detector3d/detector3d.py:199
↓ 1 callersFunctionmake_cuda_ext
(name, module, sources, sources_cuda=[],
setup_voxel_pooling.py:12
↓ 1 callersFunctionmatcher_cfgs
(matcher_name, cfg)
model/modeling/detector3d/detector3d.py:458
↓ 1 callersFunctionmesh_cuboid
(box3d=None, R=None, color=None)
model/util/math_util.py:762
↓ 1 callersFunctionneck_cfgs
(neck_name, cfg)
model/modeling/detector3d/detector3d.py:319
↓ 1 callersFunctionprepare_mm_omni3d_ar
(ar_root, omni3d_root)
data_preparation/arkitscenes_prepare.py:11
↓ 1 callersFunctionprepare_mm_omni3d_hypersim2
(hypersim_root, hypersim_meta_root, omni3d_root, MM_omni3d_root)
data_preparation/hypersim_prepare.py:130
↓ 1 callersFunctionprepare_mm_omni3d_kitti
(kitti_root, omni3d_root)
data_preparation/kitti_prepare.py:10
↓ 1 callersFunctionprepare_mm_omni3d_nuscenes
(nus_root, omni3d_root)
data_preparation/nuscenes_prepare.py:12
↓ 1 callersFunctionprepare_mm_omni3d_objectron
(omni3d_root, objectron_url, objectron_meta_root)
data_preparation/objectron_prepare.py:13
↓ 1 callersFunctionprepare_mm_omni3d_sunrgbd
(sunrgbd_root, omni3d_root)
data_preparation/sunrgbd_prepare.py:10
↓ 1 callersMethodpreprocess_image
Normalize, pad and batch the input images.
model/modeling/meta_arch/MM_UniMODE.py:48
↓ 1 callersFunctionproject_points
(points, projection_matrix, view_matrix, width, height)
data_preparation/objectron_prepare.py:66
↓ 1 callersFunctionpts_backbone_cfgs
(pts_backbone_name, cfg)
model/modeling/detector3d/detector3d.py:398
↓ 1 callersFunctionpts_middle_encoders_cfgs
(pts_middle_enc_name, cfg)
model/modeling/detector3d/detector3d.py:377
↓ 1 callersFunctionpts_neck_cfgs
(pts_neck_name, cfg)
model/modeling/detector3d/detector3d.py:412
↓ 1 callersFunctionpts_voxel_encoders_cfgs
(pts_voxel_enc_name, cfg)
model/modeling/detector3d/detector3d.py:368
↓ 1 callersFunctionpts_voxel_layer_cfgs
(pts_voxel_name, cfg)
model/modeling/detector3d/detector3d.py:354
↓ 1 callersFunctionreduce_sum
(tensor)
model/util/torch_dist.py:32
↓ 1 callersMethodremove_instance_out_range
(self, batched_inputs)
model/modeling/detector3d/detector3d.py:235
↓ 1 callersMethodremove_invalid_gts
Description: Remove the labels the class IDs of which are -1.
model/modeling/meta_arch/MM_UniMODE.py:58
↓ 1 callersFunctionrender_depth_map
(K, box3d, pose, width, height, device=None)
model/util/math_util.py:708
↓ 1 callersMethodreset_parameters
(self)
model/modeling/detector3d/detector3d.py:992
↓ 1 callersFunctionrotate_half
(x)
model/modeling/backbone/eva.py:106
↓ 1 callersMethodsave_predictions
Saves the predictions from dataset_name to disk, in a self.output_folder. Args: dataset_name (str): the dataset split na
model/evaluation/omni3d_evaluation.py:283
↓ 1 callersMethodsetDet2DParams
(self)
model/evaluation/omni3d_evaluation.py:1034
↓ 1 callersMethodsetDet3DParams
(self)
model/evaluation/omni3d_evaluation.py:1058
↓ 1 callersFunctionsetup_for_distributed
This function disables printing when not in master process
model/util/default.py:66
↓ 1 callersFunctionsigmoid_hm
(hm_features)
model/modeling/detector3d/center_head.py:215
↓ 1 callersMethodsummarize_all
Report collective metrics when possible for the the Omni3D dataset. This uses pre-computed evaluation results from each dataset,
model/evaluation/omni3d_evaluation.py:383
↓ 1 callersMethodtrain
Convert the model into training mode while keep normalization layer freezed.
model/modeling/backbone/vovnet.py:398
↓ 1 callersFunctiontransform_instance_annotations
(annotation, transforms, *, K)
model/data/dataset_mapper.py:254
↓ 1 callersMethodtransform_intrinsics
(self, images, batched_inputs)
model/modeling/detector3d/detector3d.py:210
↓ 1 callersFunctiontransformer_cfgs
(transformer_name, cfg)
model/modeling/detector3d/detector3d.py:426
↓ 1 callersMethodwith_pos_embed
(tensor, pos)
model/modeling/detector3d/deformable_detr.py:179
FunctionPYBIND11_MODULE
model/deformable_ops/src/vision.cpp:13
FunctionPYBIND11_MODULE
voxel_pooling/src/voxel_pooling_train_forward.cpp:90
FunctionR_from_allocentric
Convert a rotation matrix or series of rotation matrices to egocentric representation given a 2D location (u, v) in pixels. When u or v
model/util/math_util.py:652
FunctionR_to_allocentric
Convert a rotation matrix or series of rotation matrices to allocentric representation given a 2D location (u, v) in pixels. When u or v
model/util/math_util.py:596
Method__call__
Call function to perform photometric distortion on images. Args: results (dict): Result dict from loading pipeline. Return
model/data/augmentation.py:45
Method__call__
(self, dataset_dict)
model/data/dataset_mapper.py:116
Method__call__
(self, key)
model/util/util.py:315
Method__call__
Args: bbox_pred (Tensor): Predicted boxes with normalized coordinates (cx, cy, w, h), which are all in range [0,
model/util/match_cost.py:17
Method__init__
(self, in_features, hidden_features=None, out_features=None, act_layer=nn.SiLU, drop=0., norm
model/modeling/backbone/eva_det.py:43
← previousnext →101–200 of 454, ranked by callers