MCPcopy Create free account

hub / github.com/HavenFeng/St4RTrack / functions

Functions1,146 in github.com/HavenFeng/St4RTrack

⨍ Functions 1,146◇ Types & classes 143
↓ 1 callersFunctionget_args_parser
()
dust3r/demo.py:41
↓ 1 callersFunctionget_args_parser
()
croco/pretrain.py:37
↓ 1 callersFunctionget_args_parser
()
croco/stereoflow/train.py:37
↓ 1 callersFunctionget_args_parser
()
croco/stereoflow/test.py:30
↓ 1 callersFunctionget_batches
Creates pairs of views for batch processing. The first item in batch_data is the 'base'. We pair it with every item (including itsel
dust3r/inference.py:92
↓ 1 callersMethodget_conf
(self, mode=None)
dust3r/cloud_opt/base_opt.py:278
↓ 1 callersMethodget_cos_sin
(self, D, seq_len, device, dtype)
croco/models/pos_embed.py:115
↓ 1 callersFunctionget_data_loader
(dataset, batch_size, num_workers=8, shuffle=True, drop_last=True, pin_mem=True)
dust3r/datasets/__init__.py:26
↓ 1 callersFunctionget_depth
get depth from reference frame disparity and camera intrinsics
dust3r/datasets/spring_dataset.py:18
↓ 1 callersMethodget_depth_head2
Transforms the 3D points (pts3d_in_other_view) in pred2 to the camera coordinate system using the predicted extrinsic parameters and
dust3r/camera_solver.py:269
↓ 1 callersMethodget_depth_head2_gt
Transforms the 3D points (pts3d_in_other_view) in pred2 to the camera coordinate system using the predicted extrinsic parameters and
dust3r/camera_solver.py:303
↓ 1 callersFunctionget_dynamic_mask_from_pairviewer
get the dynamic mask from the pairviewer
dust3r/demo.py:135
↓ 1 callersMethodget_flow_inconsistency_tensor
(self, base_coord, flow_1_2, flow_2_1)
dust3r/utils/goem_opt.py:600
↓ 1 callersMethodget_focals
(self)
dust3r/cloud_opt/oneref_viewer.py:219
↓ 1 callersFunctionget_gaussian_kernel_2d
(channels, kernel_size=3, sigma=2.0, mid_one=False)
dust3r/utils/po_utils/basic.py:347
↓ 1 callersFunctionget_grad_norm_
(parameters, norm_type: float = 2.0)
croco/utils/misc.py:281
↓ 1 callersFunctionget_imshapes
(edges, pred_i, pred_j)
dust3r/cloud_opt/commons.py:38
↓ 1 callersFunctionget_inference_output
Get the inference output for a list of images
dust3r/track_eval.py:85
↓ 1 callersMethodget_init_conf
(self, mode=None)
dust3r/cloud_opt/base_opt.py:282
↓ 1 callersMethodget_intrinsics
(self)
dust3r/cloud_opt/oneref_viewer.py:222
↓ 1 callersMethodget_known_focal_mask
(self)
dust3r/cloud_opt/pair_viewer.py:98
↓ 1 callersFunctionget_known_focals
(self)
dust3r/cloud_opt/init_im_poses.py:347
↓ 1 callersFunctionget_n_colors
(N, sequential=False)
dust3r/utils/po_utils/improc.py:449
↓ 1 callersFunctionget_normal_loss
Compute geometric loss: 1) Build adjacency matrix based on GT point cloud 2) Estimate normals for both predicted and ground truth poi
dust3r/normal_loss.py:86
↓ 1 callersMethodget_oob_mask
(self, base_coord, flow_1_2)
dust3r/utils/goem_opt.py:593
↓ 1 callersFunctionget_pair_transforms
(transform_str, totensor=True, normalize=True)
croco/datasets/transforms.py:66
↓ 1 callersFunctionget_path
(idx)
croco/datasets/crops/extract_crops_from_images.py:79
↓ 1 callersFunctionget_pointwise_threshold_multiplier
Computes a per-point multiplier to turn a "pixel threshold" into a "world-space threshold" based on camera intrinsics. - gt_tracks: (T,
dust3r/track_eval_util.py:177
↓ 1 callersMethodget_principal_points
(self)
dust3r/cloud_opt/oneref_viewer.py:232
↓ 1 callersMethodget_pw_scale
(self)
dust3r/cloud_opt/base_opt.py:240
↓ 1 callersFunctionget_pytorch3d_camera
(entry_viewpoint, image_size, scale: float )
data/construct_ds_bench.py:29
↓ 1 callersFunctionget_pytorch3d_camera
(entry_viewpoint, image_size, scale: float )
dust3r/datasets/dynamic_replica.py:30
↓ 1 callersMethodget_results
(self)
croco/stereoflow/criterion.py:249
↓ 1 callersMethodget_stats
(self)
dust3r/datasets/waymo.py:41
↓ 1 callersMethodget_stats
(self)
dust3r/datasets/base/base_stereo_view_dataset.py:55
↓ 1 callersMethodget_tracks_2d
Projects the given 3D points (gt1['pts3d']) onto the image plane using the predicted extrinsic parameters (pred_poses). Retu
dust3r/camera_solver.py:186
↓ 1 callersMethodget_tracks_2d_gt
Projects the given 3D points (gt1['pts3d']) onto the image plane using the predicted extrinsic parameters (pred_poses). Retu
dust3r/camera_solver.py:230
↓ 1 callersMethodget_tum_poses
(self)
dust3r/cloud_opt/oneref_viewer.py:238
↓ 1 callersFunctionget_visibility_from_depth
(depth_maps, max_depth=5)
dust3r/track_eval_util.py:16
↓ 1 callersFunctionglobal_alignment_loop
(net, lr=0.01, niter=300, schedule='cosine', lr_min=1e-3, temporal_smoothing_weight=0, depth_map_save_dir=None
dust3r/cloud_opt/base_opt.py:456
↓ 1 callersMethodgn_step
(self, x3d, x2d, w2d, pose, camera, cost_fun)
dust3r/cloud_opt/levenberg_marquardt.py:243
↓ 1 callersFunctiongridcloud2d
(B, Y, X, norm=False, device='cuda')
dust3r/utils/po_utils/basic.py:262
↓ 1 callersFunctionhard_argmax2d
(tensor)
dust3r/utils/po_utils/basic.py:68
↓ 1 callersFunctionhat
Compute the Hat operator [1] of a batch of 3D vectors. Args: v: Batch of vectors of shape `(minibatch , 3)`. Returns: B
dust3r/utils/goem_opt.py:111
↓ 1 callersFunctionhuber_d_kernel
(s_sqrt, delta, eps: float = 1e-10)
dust3r/cloud_opt/cost_fun.py:15
↓ 1 callersFunctionhuber_kernel
(s_sqrt, delta)
dust3r/cloud_opt/cost_fun.py:8
↓ 1 callersFunctioninit_from_pts3d
(self, pts3d, im_focals, im_poses)
dust3r/cloud_opt/init_im_poses.py:106
↓ 1 callersMethodinit_grid
(self, shape, device)
dust3r/utils/goem_opt.py:43
↓ 1 callersMethodinit_grid
(self, shape, device)
dust3r/utils/goem_opt.py:581
↓ 1 callersMethodinitial_fit
(self, *args, **kwargs)
dust3r/cloud_opt/epropnp.py:62
↓ 1 callersMethodinitialize_weights
(self)
croco/models/croco.py:119
↓ 1 callersFunctioninterpolate_pos_embed
(model, checkpoint_model)
croco/models/pos_embed.py:78
↓ 1 callersMethodintrinsics_solver
(self)
dust3r/cloud_opt/oneref_viewer.py:171
↓ 1 callersFunctionintrinsics_to_pixel_space
Convert camera intrinsics to pixel space. Args: intrinsics (np.ndarray): Camera intrinsics in normalized space (3x3 matrix).
dust3r/datasets/custom.py:20
↓ 1 callersFunctionis_good_type
returns (is_good, err_msg)
dust3r/datasets/base/base_stereo_view_dataset.py:186
↓ 1 callersFunctionis_main_process
()
croco/utils/misc.py:219
↓ 1 callersMethodis_other_viewpoint_overlapping
Check if a viewpoint is valid and overlaps significantly with a reference one.
croco/datasets/habitat_sim/multiview_habitat_sim_generator.py:267
↓ 1 callersFunctionis_symmetrized
(gt1, gt2)
dust3r/utils/misc.py:58
↓ 1 callersFunctionlist_replica_cad_baked_lighting_scenes
(base_output_dir, base_path=SCENES_DATASET["replica_cad_baked_lighting"])
croco/datasets/habitat_sim/paths.py:42
↓ 1 callersFunctionlist_replica_scenes
(base_output_dir, base_path)
croco/datasets/habitat_sim/paths.py:56
↓ 1 callersFunctionlist_scenes_available
(base_output_dir, scenes_dataset_paths=SCENES_DATASET)
croco/datasets/habitat_sim/paths.py:103
↓ 1 callersFunctionload_16big_png_depth
(depth_png)
data/construct_ds_bench.py:18
↓ 1 callersFunctionload_16big_png_depth
(depth_png)
dust3r/datasets/dynamic_replica.py:118
↓ 1 callersFunctionload_crop_file
(path)
croco/datasets/crops/extract_crops_from_images.py:56
↓ 1 callersFunctionload_hf_model
Load the HuggingFace model with fresh download
infer.py:79
↓ 1 callersFunctionload_hf_model
Load the HuggingFace model with fresh download
dust3r/training.py:183
↓ 1 callersFunctionload_image
(path)
croco/datasets/crops/extract_crops_from_images.py:99
↓ 1 callersFunctionload_model
(model_path, device, verbose=True)
dust3r/model_seq.py:25
↓ 1 callersFunctionload_model
(args, device)
dust3r/training.py:152
↓ 1 callersFunctionload_model
(model_path, device, verbose=False)
dust3r/model.py:28
↓ 1 callersFunctionload_npz_data_recon
(npz_path, num_frames=None, normalize_cam=True)
dust3r/datasets/tapvid3d.py:495
↓ 1 callersFunctionload_pairs_from_cache_file
(fname, root='')
croco/datasets/pairs_dataset.py:13
↓ 1 callersFunctionload_pairs_from_list_file
(fname, root='')
croco/datasets/pairs_dataset.py:20
↓ 1 callersFunctionload_replica_traj
(gt_file)
dust3r/utils/vo_eval.py:40
↓ 1 callersFunctionload_sintel_traj
(gt_file)
dust3r/utils/vo_eval.py:69
↓ 1 callersMethodload_state_dict
(self, ckpt, **kw)
dust3r/model_seq.py:99
↓ 1 callersFunctionload_traj
Read trajectory format. Return in TUM-RGBD format. Returns: traj_tum (N, 7): camera to world poses in (x,y,z,qx,qy,qz,qw) timestam
dust3r/utils/vo_eval.py:97
↓ 1 callersFunctionlook_at
Return camera pose looking at a given center point. Analogous of gluLookAt function, using OpenCV camera convention.
croco/datasets/habitat_sim/multiview_habitat_sim_generator.py:84
↓ 1 callersFunctionmain
(args)
croco/pretrain.py:73
↓ 1 callersFunctionmain
(args)
croco/stereoflow/train.py:85
↓ 1 callersFunctionmain
(args)
croco/stereoflow/test.py:128
↓ 1 callersFunctionmain
(args)
croco/datasets/crops/extract_crops_from_images.py:30
↓ 1 callersFunctionmain_demo
(tmpdirname, model, device, image_size, server_name, server_port, silent=False, args=None)
dust3r/demo.py:339
↓ 1 callersFunctionmake_batch_symmetric
(batch)
dust3r/inference.py:38
↓ 1 callersFunctionmake_color_wheel
Generate color wheel according Middlebury color code :return: Color wheel
dust3r/utils/flow_vis.py:176
↓ 1 callersFunctionmake_pairs
(imgs, scene_graph='complete', prefilter=None, symmetrize=True, force_symmetrize=False)
dust3r/image_pairs.py:11
↓ 1 callersMethodmake_sampler
(self, batch_size, shuffle=True, world_size=1, rank=0, drop_last=True)
dust3r/datasets/base/easy_dataset.py:38
↓ 1 callersFunctionmake_scratch
(in_shape, out_shape, groups=1, expand=False)
croco/models/dpt_block.py:20
↓ 1 callersMethodmask_sky
(self)
dust3r/cloud_opt/base_opt.py:420
↓ 1 callersFunctionmeshgrid3d
(B, Z, Y, X, stack=False, norm=False, device='cuda')
dust3r/utils/po_utils/basic.py:206
↓ 1 callersFunctionminimum_spanning_tree
(imshapes, edges, pred_i, pred_j, conf_i, conf_j, im_conf, min_conf_thr, device, has
dust3r/cloud_opt/init_im_poses.py:156
↓ 1 callersMethodmonte_carlo_forward
Monte Carlo PnP forward. Returns weighted pose samples drawn from the probability distribution of pose defined by the correspondences
dust3r/cloud_opt/epropnp.py:80
↓ 1 callersFunctionmonte_carlo_pose_loss
Args: pose_sample_logweights: Shape (mc_samples, num_obj) cost_target: Shape (num_obj, ) Returns: Tensor: Shape (num
dust3r/cloud_opt/oneref_viewer.py:67
↓ 1 callersFunctionnormal_consistency_loss
Compute normal consistency loss: Mean of 1 - |dot(n_pred, n_gt)| for each point. Args: pred_normals (torch.Tensor): (B, N, 3)
dust3r/normal_loss.py:69
↓ 1 callersFunctionnormalize_boxlist2d
(boxlist2d, H, W)
dust3r/utils/po_utils/basic.py:321
↓ 1 callersFunctionnormalize_grid2d
(grid_y, grid_x, Y, X, clamp_extreme=True)
dust3r/utils/po_utils/basic.py:238
↓ 1 callersFunctionnormalize_grid3d
(grid_z, grid_y, grid_x, Z, Y, X, clamp_extreme=True)
dust3r/utils/po_utils/basic.py:249
↓ 1 callersFunctionnormalize_single
(d)
dust3r/utils/po_utils/basic.py:51
↓ 1 callersFunctionopencv_to_colmap_intrinsics
Modify camera intrinsics to follow a different convention. Coordinates of the center of the top-left pixels are by default: - (0.5, 0.5)
dust3r/utils/geometry.py:281
↓ 1 callersFunctionpair
(t)
croco/models/dpt_block.py:17
← previousnext →401–500 of 1,146, ranked by callers