Functions — github.com/Lijiaxin0111/Open-d4rt

hub / github.com/Lijiaxin0111/Open-d4rt / functions

Functions626 in github.com/Lijiaxin0111/Open-d4rt

⨍ Functions 626◇ Types & classes 62

↓ 1 callersMethod_getitem_dslr

(self, scene: _DSLRScene, start: int, idxs: list[int] | None = None)

src/data/scannet_raw_dataset.py:594

↓ 1 callersMethod_getitem_iphone

(self, scene: _IphoneScene, start: int, idxs: list[int] | None = None)

src/data/scannet_raw_dataset.py:504

↓ 1 callersFunction_git_commit

(cwd: Path)

train.py:100

↓ 1 callersFunction_grid_query_points

( width: int, height: int, cols: int, rows: int, margin_ratio: float, max_points: int,

infer_track_3d.py:68

↓ 1 callersMethod_has_usable_depth

(self, frames: list[_FrameMeta])

src/data/co3d_raw_dataset.py:399

↓ 1 callersMethod_in_split

(self, scene_name: str)

src/data/blendermvs_raw_dataset.py:260

↓ 1 callersMethod_in_split

(self, clip_name: str)

src/data/mvs_synth_raw_dataset.py:198

↓ 1 callersFunction_infer_point_cloud_ref0

( *, model: torch.nn.Module, video_model_rgb: np.ndarray, point_query_uv_norm: np.ndarray,

vis/build_like_demo.py:828

↓ 1 callersFunction_infer_regular_grid_shape

(query_uv_px: np.ndarray)

vis/build_like_demo.py:123

↓ 1 callersFunction_infinite

(loader: Iterable[Any])

src/engine/trainer.py:174

↓ 1 callersFunction_init_distributed

Initialize DDP from torchrun env vars. Returns (rank, local_rank, world_size).

train.py:37

↓ 1 callersFunction_is_higher_better

(sort_key: str)

scripts/eval_worldtrack/batch_eval_worldtrack_step_ckpts.py:543

↓ 1 callersFunction_is_nan_scalar_placeholder

(arr: np.ndarray | None)

src/data/pointodyssey_raw_dataset.py:55

↓ 1 callersFunction_json_scalar

(value: Any)

src/data_preprocess/prepare_kubric_full_preprocessed.py:105

↓ 1 callersMethod_launch_worldtrack_step_eval

(self, step_ckpt: Path)

src/engine/trainer.py:431

↓ 1 callersFunction_lighten_colors

(colors: np.ndarray, amount: float = 0.4)

vis/serve_demo_viser.py:148

↓ 1 callersMethod_load_camera_index

(self, split_dir: Path, split: str)

src/data/dynamic_replica_raw_dataset.py:402

↓ 1 callersMethod_load_category_scenes

( self, category: str, category_dir: Path, wanted_frames: dict[str, set[int]],

src/data/co3d_raw_dataset.py:298

↓ 1 callersFunction_load_demo

(root: Path)

vis/serve_demo_viser.py:111

↓ 1 callersFunction_load_dslr_scene

(scene_root: Path)

src/data/scannet_raw_dataset.py:319

↓ 1 callersFunction_load_iphone_scene

(scene_root: Path)

src/data/scannet_raw_dataset.py:370

↓ 1 callersFunction_load_kubric_camera_sequence

( *, tfds_root: Path, sample_json_path: Path, expected_num_frames: int, )

vis/build_like_demo.py:364

↓ 1 callersFunction_load_kubric_sample

( *, tfds_root: Path, sample_json_path: Path, )

vis/build_like_demo.py:412

↓ 1 callersMethod_load_pretrained_encoder_weights

(self, pretrained_cfg: Any)

src/model/d4rt.py:286

↓ 1 callersMethod_load_scene

(self, scene_dir: Path, scene_id: str)

src/data/blendermvs_raw_dataset.py:277

↓ 1 callersMethod_load_scene

(self, clip_dir: Path)

src/data/mvs_synth_raw_dataset.py:215

↓ 1 callersMethod_load_scene_cache

(self, scene: _Scene)

src/data/pointodyssey_raw_dataset.py:168

↓ 1 callersFunction_loader_bool

(split: str, cfg: Any, key: str, default: bool)

src/data/builder.py:701

↓ 1 callersMethod_log_case

(self, batch: dict[str, Any], outputs: dict[str, Any], split: str, step: int, case_idx: int, batch_index: int)

src/vis/training_visualizer.py:395

↓ 1 callersFunction_make_normalized_uv_grid

(grid_size: int)

vis/build_like_demo.py:116

↓ 1 callersFunction_make_source_pose

()

vis/serve_demo_viser.py:163

↓ 1 callersFunction_make_tqdm

(*, total: int, desc: str)

scripts/eval_worldtrack/batch_eval_worldtrack_step_ckpts.py:66

↓ 1 callersMethod_merge_dict_inplace

(self, target: dict[str, Any], update: dict[str, Any])

src/engine/trainer.py:564

↓ 1 callersFunction_metrics_for_sequence

( gt_tracks_world: np.ndarray, pred_tracks_ref0: np.ndarray, compute_dyn: bool = True, )

eval_track3d_in_worldtrack.py:295

↓ 1 callersFunction_ned_pose_to_t_wc

Convert TartanAir NED pose (tx ty tz qx qy qz qw) to T_wc in standard camera frame. The pose describes camera-to-world in the NED coordinate syst

src/data/tartanair_raw_dataset.py:70

↓ 1 callersMethod_next_train_sample

(self)

src/data/kubric_full_robust_dataset.py:413

↓ 1 callersFunction_normalize_attention_pattern

(pattern: str | None)

src/model/encoder.py:15

↓ 1 callersFunction_normalize_manifest_arg

(value: str | None)

src/data/builder.py:670

↓ 1 callersFunction_normalize_sample_meta

Keep only canonical keys in ``sample["meta"]``.

src/data/mixture_dataset.py:24

↓ 1 callersFunction_parse_cam

(path: Path)

src/data/blendermvs_raw_dataset.py:103

↓ 1 callersFunction_parse_colmap_camera

(cameras_path: Path)

src/data/scannet_raw_dataset.py:102

↓ 1 callersFunction_parse_colmap_images

(images_path: Path, image_dir: Path)

src/data/scannet_raw_dataset.py:124

↓ 1 callersFunction_parse_colmap_points

(points_path: Path, needed_ids: set[int])

src/data/scannet_raw_dataset.py:164

↓ 1 callersFunction_parse_extrinsics

(path: Path)

src/data/virtual_kitti2_raw_dataset.py:377

↓ 1 callersFunction_parse_intrinsics

(path: Path)

src/data/virtual_kitti2_raw_dataset.py:353

↓ 1 callersFunction_parse_iphone_pose_intrinsic

(path: Path)

src/data/scannet_raw_dataset.py:182

↓ 1 callersFunction_parse_override_value

(raw: str)

src/core/config.py:79

↓ 1 callersFunction_parse_pose

(path: Path)

src/data/mvs_synth_raw_dataset.py:73

↓ 1 callersFunction_parse_steps

(raw_steps: str)

scripts/eval_worldtrack/batch_eval_worldtrack_step_ckpts.py:187

↓ 1 callersFunction_plot_3d_on_image

(image: np.ndarray, gt_xy: np.ndarray, errors_m: np.ndarray, title: str)

src/vis/training_visualizer.py:266

↓ 1 callersFunction_plot_3d_pairs

(gt_xyz: np.ndarray, pred_xyz: np.ndarray, errors_m: np.ndarray, title: str, max_lines: int)

src/vis/training_visualizer.py:291

↓ 1 callersFunction_portable_path

(path: str | Path, base_dir: Path = REPO_ROOT)

vis/build_like_demo_for_worldtrack.py:106

↓ 1 callersFunction_pred_camera_translation_already_aligned

(meta: dict[str, Any])

vis/serve_demo_viser.py:189

↓ 1 callersFunction_predict_camera_branches

( *, model: torch.nn.Module, video_model_rgb: np.ndarray, image_hw: tuple[int, int], camer

vis/build_like_demo.py:1049

↓ 1 callersFunction_prepare_init_state_dict

Filter init weights and resize learned timestep embeddings when clip length grows.

train.py:202

↓ 1 callersMethod_project_aspect_ratio_token

( self, video: torch.Tensor, aspect_ratio: torch.Tensor | None, )

src/model/d4rt.py:231

↓ 1 callersFunction_project_point

(k: np.ndarray, t_cw: np.ndarray, p_world: np.ndarray)

vis/build_like_demo.py:468

↓ 1 callersFunction_project_points_to_video_frame

(camera_pov_points3d: np.ndarray, camera_intrinsics: np.ndarray)

eval_track3d_in_worldtrack.py:46

↓ 1 callersFunction_project_points_to_video_frame

(camera_pov_points3d: np.ndarray, camera_intrinsics: np.ndarray)

vis/build_like_demo_for_worldtrack.py:95

↓ 1 callersMethod_project_pred_xyz_to_tgt_uv

( self, *, pred_xyz_tcam: torch.Tensor, batch: dict[str, Any], )

src/losses/d4rt_loss.py:106

↓ 1 callersFunction_project_pred_xyz_to_tgt_uv_np

( *, pred_xyz_tcam: np.ndarray, k_seq: np.ndarray, t_wc_seq: np.ndarray, camera_valid: np.

src/vis/training_visualizer.py:219

↓ 1 callersFunction_quat_to_rot

(qw: float, qx: float, qy: float, qz: float)

src/data/scannet_raw_dataset.py:79

↓ 1 callersFunction_quat_to_rotation

Convert quaternion (x, y, z, w) to 3x3 rotation matrix.

src/data/tartanair_raw_dataset.py:54

↓ 1 callersFunction_quat_wxyz_to_rot

(quat_wxyz: np.ndarray)

vis/build_like_demo.py:342

↓ 1 callersFunction_quat_wxyz_to_rot

(q_wxyz: np.ndarray)

src/data/kubric_full_robust_dataset.py:37

↓ 1 callersFunction_rank_summary_rows

(rows: list[dict[str, Any]], sort_key: str)

scripts/eval_worldtrack/batch_eval_worldtrack_step_ckpts.py:343

↓ 1 callersFunction_read_dataset_info

(tfds_dir: Path)

src/data_preprocess/prepare_kubric_full_preprocessed.py:242

↓ 1 callersFunction_read_depth

(path: Path, width: int, height: int)

src/data/pointodyssey_raw_dataset.py:40

↓ 1 callersFunction_read_depth_exr

(path: Path, width: int, height: int)

src/data/mvs_synth_raw_dataset.py:58

↓ 1 callersFunction_read_depth_pfm

(path: Path, width: int, height: int)

src/data/blendermvs_raw_dataset.py:70

↓ 1 callersFunction_read_depth_png

(path: Path, width: int, height: int)

src/data/virtual_kitti2_raw_dataset.py:42

↓ 1 callersFunction_read_depth_png

(path: Path, width: int, height: int)

src/data/co3d_raw_dataset.py:49

↓ 1 callersFunction_read_depth_rgba_png

Read TartanAir V2 RGBA-encoded depth PNG and return depth in meters. The 4 RGBA uint8 channels encode a float32 depth value (little-endian).

src/data/tartanair_raw_dataset.py:95

↓ 1 callersFunction_read_image

(path: Path, width: int, height: int)

src/data/scannet_raw_dataset.py:51

↓ 1 callersFunction_read_mask_png

(path: Path, width: int, height: int)

src/data/co3d_raw_dataset.py:58

↓ 1 callersFunction_read_rgb

(path: Path, width: int, height: int)

src/data/pointodyssey_raw_dataset.py:32

↓ 1 callersFunction_read_rgb

(path: Path, width: int, height: int)

src/data/blendermvs_raw_dataset.py:52

↓ 1 callersFunction_read_rgb

(path: Path, width: int, height: int)

src/data/virtual_kitti2_raw_dataset.py:34

↓ 1 callersFunction_read_rgb

(path: Path, width: int, height: int)

src/data/tartanair_raw_dataset.py:87

↓ 1 callersFunction_read_rgb

(path: Path, width: int, height: int)

src/data/mvs_synth_raw_dataset.py:50

↓ 1 callersFunction_read_rgb

(path: Path, width: int, height: int)

src/data/co3d_raw_dataset.py:41

↓ 1 callersFunction_read_scene_flow_png

Decode VKITTI2 forward scene flow from 16-bit RGB PNG to float32 [H,W,3] in meters. Encoding: dx = ((R * 2.0 / 65535.0) - 1.0) * 10.0 (range [-1

src/data/virtual_kitti2_raw_dataset.py:51

↓ 1 callersFunction_read_video_frames

(path: Path, frame_ids: list[int], width: int, height: int)

src/data/scannet_raw_dataset.py:59

↓ 1 callersMethod_repo_root

(self)

src/engine/trainer.py:402

↓ 1 callersFunction_resize_depth

(depth_m: np.ndarray, width: int, height: int)

src/data/scannet_raw_dataset.py:262

↓ 1 callersFunction_resize_depth

(path: Path, width: int, height: int)

src/data/dynamic_replica_raw_dataset.py:44

↓ 1 callersFunction_resize_patch_embed_weight

(src: torch.Tensor, dst_shape: torch.Size)

src/model/d4rt.py:87

↓ 1 callersFunction_resize_rgb

(path: Path, width: int, height: int)

src/data/dynamic_replica_raw_dataset.py:34

↓ 1 callersFunction_resize_timestep_embedding

Resize [T, C] learned timestep embeddings along T.

train.py:168

↓ 1 callersMethod_resize_to_target

( self, video: np.ndarray, depth_z: np.ndarray, depth_range_m: np.ndarray,

src/data/kubric_full_robust_dataset.py:527

↓ 1 callersFunction_resolve_mixture_weights

(cfg: Any, selected_sources: list[str])

src/data/builder.py:622

↓ 1 callersFunction_resolve_tfds_dir

(root: Path)

src/data/kubric_full_robust_dataset.py:109

↓ 1 callersMethod_root_tag

(self, root: Path)

src/data/blendermvs_raw_dataset.py:249

↓ 1 callersFunction_run_clip_queries_for_target_indices

( *, model: torch.nn.Module, video_clip: torch.Tensor, aspect_ratio: torch.Tensor | None,

infer_track_3d.py:301

↓ 1 callersFunction_run_full_clip_queries

( *, model: torch.nn.Module, video_clip: torch.Tensor, aspect_ratio: torch.Tensor | None,

infer_track_3d.py:235

↓ 1 callersMethod_run_reprojection_self_check

(self)

src/data/dynamic_replica_raw_dataset.py:453

↓ 1 callersFunction_sample_bool_mask_from_uv_sequence

(mask_thw: np.ndarray, uv_px: np.ndarray)

vis/build_like_demo.py:933

↓ 1 callersMethod_sample_by_index

(self, index: int)

src/data/kubric_full_robust_preprocess_dataset.py:132

↓ 1 callersMethod_sample_key

(self, scene: _Scene, frame_ids: list[int])

src/data/dynamic_replica_raw_dataset.py:586

↓ 1 callersMethod_sample_key

(self, scene: _Scene, idxs: list[int])

src/data/pointodyssey_raw_dataset.py:289

← previous next →301–400 of 626, ranked by callers