MCPcopy Create free account

hub / github.com/NVlabs/InstantSplat / functions

Functions864 in github.com/NVlabs/InstantSplat

⨍ Functions 864◇ Types & classes 85
↓ 57 callersFunctionto_numpy
(x)
dust3r/utils/device.py:42
↓ 32 callersFunctiongeotrf
Apply a geometric transformation to a list of 3-D points. H: 3x3 or 4x4 projection matrix (typically a Homography) p: numpy/torch/tuple of c
dust3r/utils/geometry.py:40
↓ 32 callersMethodwrite
(self, x)
utils/general_utils.py:118
↓ 26 callersFunctionview_name
(view, batch_index=None)
dust3r/datasets/base/base_stereo_view_dataset.py:194
↓ 19 callersFunctionwrite_next_bytes
pack and write to a binary file. :param fid: :param data: data to send, if multiple elements are sent at the same time, they should be enc
scene/colmap_loader.py:320
↓ 17 callersFunctionimread_cv2
Open an image or a depthmap with opencv-python.
dust3r/utils/image.py:31
↓ 16 callersMethodadd_pointcloud
(self, pts3d, color=(0,0,0), mask=None, denoise=False)
dust3r/viz.py:137
↓ 16 callersFunctioninv
Invert a torch or numpy matrix
dust3r/utils/geometry.py:104
↓ 16 callersMethodsave
(self, iteration)
scene/__init__.py:98
↓ 14 callersFunctionedge_str
(i, j)
dust3r/cloud_opt/commons.py:12
↓ 14 callersFunctionrgb
(ftensor, true_shape=None)
dust3r/utils/image.py:44
↓ 13 callersFunctionread_next_bytes
Read and unpack the next bytes from a binary file. :param fid: :param num_bytes: Sum of combination of {2, 4, 8}, e.g. 2, 6, 16, 30, etc.
scene/colmap_loader.py:73
↓ 12 callersFunctionvector_norm
Return length, i.e. eucledian norm, of ndarray along axis. >>> v = numpy.random.random(3) >>> n = vector_norm(v) >>> numpy.allclose(n, nu
utils/utils_poses/ATE/transformations.py:1804
↓ 11 callersMethodadd_camera
(self, pose_c2w, focal=None, color=(0, 0, 0), image=None, imsize=None, cam_size=0.03)
dust3r/viz.py:190
↓ 11 callersFunctionfocal2fov
(focal, pixels)
utils/graphics_utils.py:96
↓ 10 callersFunctionauto_cam_size
(im_poses)
dust3r/viz.py:115
↓ 10 callersFunctioninvalid_to_nans
(arr, valid_mask, ndim=999)
dust3r/utils/misc.py:103
↓ 10 callersFunctionunit_vector
Return ndarray normalized by length, i.e. eucledian norm, along axis. >>> v0 = numpy.random.random(3) >>> v1 = unit_vector(v0) >>> numpy.
utils/utils_poses/ATE/transformations.py:1843
↓ 9 callersFunctionParameterStack
(params, keys=None, is_param=None, fill=0)
dust3r/cloud_opt/optimizer.py:214
↓ 9 callersMethodshow
(self, show_pw_cams=False, show_pw_pts3d=False, cam_size=None, **kw)
dust3r/cloud_opt/base_opt.py:297
↓ 9 callersMethodsize
(self)
dust3r/datasets/utils/cropping.py:41
↓ 8 callersMethod_crop_resize_if_necessary
This function: - first downsizes the image with LANCZOS inteprolation, which is better than bilinear interpolation in
dust3r/datasets/base/base_stereo_view_dataset.py:137
↓ 8 callersMethodgetTrainCameras
(self, scale=1.0)
scene/__init__.py:102
↓ 7 callersMethod__init__
(self, mask_bg=True, *args, ROOT, **kwargs)
mast3r/datasets/__init__.py:30
↓ 7 callersFunctiondepthmap_to_absolute_camera_coordinates
Args: - depthmap (HxW array): - camera_intrinsics: a 3x3 matrix - camera_pose: a 4x3 or 4x4 cam2world matrix Returns:
dust3r/utils/geometry.py:200
↓ 7 callersMethoddevice
(self)
dust3r/cloud_opt/base_opt.py:120
↓ 7 callersMethodget_im_poses
(self)
mast3r/cloud_opt/sparse_ga.py:64
↓ 7 callersMethodget_pts3d
(self, raw=False)
dust3r/cloud_opt/base_opt.py:203
↓ 6 callersFunctionapply_log_to_norm
(xyz)
mast3r/losses.py:20
↓ 6 callersFunctioncat
(a, b)
dust3r/viz.py:322
↓ 6 callersMethodextract
(self, args)
arguments/__init__.py:40
↓ 6 callersFunctionfast_reciprocal_NNs
(pts1, pts2, subsample_or_initxy1=8, ret_xy=True, pixel_tol=0, ret_basin=False, device
mast3r/fast_nn.py:109
↓ 6 callersMethodget_focals
(self)
mast3r/cloud_opt/sparse_ga.py:58
↓ 6 callersFunctionhash_md5
(s)
mast3r/utils/misc.py:16
↓ 6 callersFunctionnormalize
Normalize a vector.
utils/utils_poses/vis_pose_utils.py:60
↓ 6 callersFunctionsRT_to_4x4
(scale, R, T, device)
dust3r/cloud_opt/init_im_poses.py:238
↓ 6 callersFunctionsave_time
(time_dir, process_name, sub_time)
utils/sfm_utils.py:43
↓ 6 callersFunctionto_cpu
(x)
dust3r/utils/device.py:43
↓ 6 callersFunctiontodevice
Transfer some variables to another device (i.e. GPU, CPU:torch, CPU:numpy). batch: list, tuple, dict of tensors or other things device: pyto
dust3r/utils/device.py:11
↓ 5 callersMethod__init__
(self, criterion, norm_mode='avg_dis', gt_scale=False, opt_fit_gt=False, sky_loss_value=2, ma
mast3r/losses.py:28
↓ 5 callersMethod_get_depthmaps
(self, TSDF_filtering_thresh=None)
mast3r/cloud_opt/tsdf_optimizer.py:27
↓ 5 callersMethod_set_pose
(self, poses, idx, R, T=None, scale=None, force=False)
dust3r/cloud_opt/base_opt.py:157
↓ 5 callersFunctionconvert3x4_4x4
:param input: (N, 3, 4) or (3, 4) torch or np :return: (N, 4, 4) or (4, 4) torch or np
utils/utils_poses/lie_group_helper.py:34
↓ 5 callersMethodcrop
(self, *args, **kwargs)
dust3r/datasets/utils/cropping.py:49
↓ 5 callersFunctiondepthmap_to_pts3d
Args: - depthmap (BxHxW array): - pseudo_focal: [B,H,W] ; [B,2,H,W] or [B,1,H,W] Returns: pointmap of absolute coordi
dust3r/utils/geometry.py:114
↓ 5 callersFunctionfov2focal
(fov, pixels)
utils/graphics_utils.py:93
↓ 5 callersMethodload_state_dict
(self, ckpt, **kw)
dust3r/model.py:90
↓ 5 callersFunctionmkdir_for
(f)
mast3r/utils/misc.py:11
↓ 5 callersFunctionquaternion_matrix
Return homogeneous rotation matrix from quaternion. >>> R = quaternion_matrix([0.06146124, 0, 0, 0.99810947]) >>> numpy.allclose(R, rotation_
utils/utils_poses/ATE/transformations.py:1411
↓ 5 callersFunctionrender
Render the scene. Background tensor (bg_color) must be on GPU!
gaussian_renderer/__init__.py:23
↓ 5 callersMethodstep
Performs a single optimization step.
scene/per_point_adam.py:34
↓ 5 callersFunctionxy_grid
Output a (H,W,2) array of int32 with output[j,i,0] = i + origin[0] output[j,i,1] = j + origin[1]
dust3r/utils/geometry.py:15
↓ 4 callersFunctionNoGradParamDict
(x)
dust3r/cloud_opt/commons.py:28
↓ 4 callersFunctionSO3_to_quat
:param R: (N, 3, 3) or (3, 3) np :return: (N, 4, ) or (4, ) np
utils/utils_poses/lie_group_helper.py:14
↓ 4 callersMethod__init__
(self, criterion, norm_mode='avg_dis', gt_scale=False)
dust3r/losses.py:153
↓ 4 callersMethod__init__
(self)
lpipsPyTorch/modules/networks.py:89
↓ 4 callersFunction_cell_size
(cell2)
mast3r/utils/coarse_to_fine.py:43
↓ 4 callersMethod_downstream_head
(self, head_num, decout, img_shape)
dust3r/model.py:192
↓ 4 callersMethod_get_msk_indices
(self, msk)
dust3r/cloud_opt/optimizer.py:103
↓ 4 callersFunction_resize_pil_image
(img, long_edge_size)
dust3r/utils/image.py:63
↓ 4 callersMethodadd_cameras
(self, poses, focals=None, images=None, imsizes=None, colors=None, **kw)
dust3r/viz.py:202
↓ 4 callersFunctionadd_scene_cam
(scene, pose_c2w, edge_color, image=None, focal=None, imsize=None, screen_width=0.03, marke
dust3r/viz.py:246
↓ 4 callersFunctionargmin
(X, dim)
mast3r/fast_nn.py:33
↓ 4 callersFunctiongamma_loss
(gamma, mul=1, offset=None, clip=np.inf)
mast3r/cloud_opt/utils/losses.py:19
↓ 4 callersFunctiongetWorld2View2
(R, t, translate=np.array([.0, .0, .0]), scale=1.0)
utils/graphics_utils.py:38
↓ 4 callersMethodget_conf_log
(self, x)
dust3r/losses.py:217
↓ 4 callersFunctionget_expon_lr_func
Copied from Plenoxels Continuous learning rate decay function. Adapted from JaxNeRF The returned rate is lr_init when step=0 and lr_fina
utils/general_utils.py:29
↓ 4 callersFunctionget_joint_pointcloud_center_scale
(pts1, pts2, valid_mask1=None, valid_mask2=None, z_only=False, center=True)
dust3r/utils/geometry.py:328
↓ 4 callersFunctionget_joint_pointcloud_depth
(z1, z2, valid_mask1, valid_mask2=None, quantile=0.5)
dust3r/utils/geometry.py:313
↓ 4 callersMethodget_masks
(self)
dust3r/cloud_opt/base_opt.py:197
↓ 4 callersFunctionget_pred_pts3d
(gt, pred, use_pose=False)
dust3r/inference.py:81
↓ 4 callersMethodget_principal_points
(self)
mast3r/cloud_opt/sparse_ga.py:61
↓ 4 callersFunctionmake_pairs
(imgs, scene_graph='complete', prefilter=None, symmetrize=True)
dust3r/image_pairs.py:11
↓ 4 callersFunctionnormalize_pointcloud
renorm pointmaps pts1, pts2 with norm_mode
dust3r/utils/geometry.py:249
↓ 4 callersMethodquery
(self, queries, k=1, **kw)
mast3r/fast_nn.py:78
↓ 4 callersMethodresize
(self, *args, **kwargs)
dust3r/datasets/utils/cropping.py:46
↓ 4 callersFunctionsel
(x, kept)
dust3r/image_pairs.py:71
↓ 4 callersMethodset_epoch
(self, epoch)
dust3r/datasets/base/easy_dataset.py:132
↓ 4 callersMethodshow
(self, point_size=2)
dust3r/viz.py:208
↓ 4 callersFunctiontranspose_to_landscape
Predict in the correct aspect-ratio, then transpose the result in landscape and stack everything back together.
dust3r/utils/misc.py:54
↓ 4 callersFunctionunravel_xy
(pos, shape)
mast3r/datasets/utils/cropping.py:102
↓ 4 callersFunctionviewmatrix
Construct lookat view matrix.
utils/pose_utils.py:221
↓ 3 callersMethod__init__
(self, parser: ArgumentParser, name : str, fill_none = False)
arguments/__init__.py:20
↓ 3 callersMethod_backproj_pts3d
Backprojection operation: from image depths to 3D points
mast3r/cloud_opt/tsdf_optimizer.py:183
↓ 3 callersMethod_check_all_imgs_are_selected
(self, msk)
dust3r/cloud_opt/optimizer.py:63
↓ 3 callersMethod_encode_image
(self, image, true_shape)
dust3r/model.py:127
↓ 3 callersMethod_encode_image_pairs
(self, img1, img2, true_shape1, true_shape2)
dust3r/model.py:141
↓ 3 callersFunction_getIndices
(n_aligned, total_n)
utils/utils_poses/ATE/align_utils.py:18
↓ 3 callersMethod_get_msk_indices
(self, msk)
dust3r/cloud_opt/modular_optimizer.py:73
↓ 3 callersMethod_get_poses
(self, poses)
dust3r/cloud_opt/base_opt.py:150
↓ 3 callersMethod_no_grad
(self, tensor)
dust3r/cloud_opt/optimizer.py:118
↓ 3 callersMethod_no_grad
(self, tensor)
dust3r/cloud_opt/modular_optimizer.py:70
↓ 3 callersFunction_norm_windows
(cell2, H2, W2, forced_resolution=None)
mast3r/utils/coarse_to_fine.py:50
↓ 3 callersFunctionanchor_depth_offsets
(canon_depth, pixels, subsample=8)
mast3r/cloud_opt/sparse_ga.py:858
↓ 3 callersFunctionarcball_constrain_to_axis
Return sphere point perpendicular to axis.
utils/utils_poses/ATE/transformations.py:1753
↓ 3 callersFunctionbfs
(tree, start_node)
mast3r/cloud_opt/sparse_ga.py:984
↓ 3 callersFunctionclean_pointcloud
Method: 1) express all 3d points in each camera coordinate frame 2) if they're in front of a depthmap --> then lower their confidence
dust3r/cloud_opt/base_opt.py:370
↓ 3 callersMethodcompute_global_alignment
(self, init=None, niter_PnP=10, focal_avg=False, known_focal=None, **kw)
dust3r/cloud_opt/base_opt.py:276
↓ 3 callersFunctionconvert3x4_4x4
:param input: (N, 3, 4) or (3, 4) torch or np :return: (N, 4, 4) or (4, 4) torch or np
utils/utils_poses/vis_pose_utils.py:149
↓ 3 callersFunctionestimate_focal
(pts3d_i, pp=None)
dust3r/cloud_opt/init_im_poses.py:245
next →1–100 of 864, ranked by callers