MCPcopy Create free account

hub / github.com/facebookresearch/sam-3d-objects / functions

Functions784 in github.com/facebookresearch/sam-3d-objects

Methodloss
(self, x1: torch.Tensor, *args_conditionals, **kwargs_conditionals)
sam3d_objects/model/backbone/generator/flow_matching/model.py:158
Methodloss
Compute shortcut model loss with mixed flow matching and self-consistency objectives
sam3d_objects/model/backbone/generator/shortcut/model.py:300
Functionmake_dict
(**kwargs)
sam3d_objects/config/utils.py:165
Functionmake_list_from_kwargs
(**kwargs)
sam3d_objects/config/utils.py:156
Functionmake_scene
(*outputs, in_place=False)
notebook/inference.py:255
Functionmake_set
(*args)
sam3d_objects/config/utils.py:148
Functionmake_string
(value)
sam3d_objects/config/utils.py:161
Functionmake_tuple
(*args)
sam3d_objects/config/utils.py:152
Functionmake_video
Creates a video by updating the camera view location and saving snapshots. Args: scene: A Plotly Figure object, typically created wi
sam3d_objects/utils/visualization/plotly/save_scene.py:19
Functionmake_zeros
(x)
sam3d_objects/model/backbone/generator/classifier_free_guidance.py:11
Functionmodulate
(x, shift, scale)
sam3d_objects/model/backbone/tdfy_dit/modules/utils.py:71
Methodnormalize
(self, pointmap: torch.Tensor, mask: torch.Tensor, scale: Optional[torch.Tensor] = None, shift: Option
sam3d_objects/data/dataset/tdfy/img_and_mask_transforms.py:503
Methodnormalize
(self, pointmap: torch.Tensor, mask: torch.Tensor, scale: Optional[torch.Tensor] = None, shift: Option
sam3d_objects/data/dataset/tdfy/img_and_mask_transforms.py:586
Methodnormalize
(self, pointmap: torch.Tensor, mask: torch.Tensor, scale: Optional[torch.Tensor] = Non
sam3d_objects/data/dataset/tdfy/img_and_mask_transforms.py:648
Methodnormalize
(self, pointmap: torch.Tensor, mask: torch.Tensor, scale: Optional[torch.Tensor] = None, shift: Option
sam3d_objects/data/dataset/tdfy/img_and_mask_transforms.py:688
Methodnum_leaf_nodes
(self)
sam3d_objects/model/backbone/tdfy_dit/representations/octree/octree_dfs.py:198
Methodnum_non_leaf_nodes
(self)
sam3d_objects/model/backbone/tdfy_dit/representations/octree/octree_dfs.py:194
Methodoccupancy
(self)
sam3d_objects/model/backbone/tdfy_dit/representations/octree/octree_dfs.py:206
Functionpad_to_square_centered
(image, value=0, pointmap=None)
sam3d_objects/data/dataset/tdfy/img_processing.py:110
Functionperturb_mask_boundary
Applies data augmentation to the mask by randomly eroding or dilating the mask. Args: image: (C, H, W) float32 [0, 1] tensor.
sam3d_objects/data/dataset/tdfy/img_and_mask_transforms.py:840
Functionperturb_mask_translation
Applies data augmentation to the mask by randomly translating the mask. Args: image: (C, H, W) float32 [0, 1] tensor. mask:
sam3d_objects/data/dataset/tdfy/img_and_mask_transforms.py:807
Methodplot_multi_objects
( pose_targets, mask_names=None, pointmap=None, pointmap_colors=None,
sam3d_objects/utils/visualization/scene_visualizer.py:68
Methodplot_scene
( points_local: torch.Tensor, instance_quaternions_l2c: torch.Tensor, instance_positio
sam3d_objects/utils/visualization/scene_visualizer.py:26
Functionpreprocess_img
( x, mask=None, img_target_shape=224, mask_target_shape=256, normalize=False, )
sam3d_objects/data/dataset/tdfy/img_processing.py:64
Functionpreprocess_mesh
(mesh: trimesh.Trimesh)
sam3d_objects/pipeline/inference_utils.py:763
Functionprint_stats
(tensor, label=None)
sam3d_objects/data/utils.py:225
Functionprocess_and_save_alignment
Complete pipeline for processing 3DB alignment and saving the result. Args: mesh_path: Path to input 3DB mesh (.ply) mask_pa
notebook/mesh_alignment.py:240
Methodrandom_generator
(self)
sam3d_objects/model/backbone/generator/base.py:30
Functionrandom_pad
(img, mask=None, max_ratio=0.0, pointmap=None)
sam3d_objects/data/dataset/tdfy/img_processing.py:154
Functionready_gaussian_for_video_rendering
(scene_gs, in_place=False, fix_alignment=False)
notebook/inference.py:190
Methodrecurse
(data)
sam3d_objects/config/utils.py:111
Functionrecursive_fn
(b)
sam3d_objects/pipeline/inference_pipeline_pointmap.py:45
Functionremove_prefix_state_dict_fn
(prefix: str)
sam3d_objects/model/io.py:85
Functionrename_checkpoint_weights_using_suffix_matching
( checkpoint_path_in, checkpoint_path_out, model: torch.nn.Module, strict: bool = True, ke
sam3d_objects/model/io.py:18
Functionrender_gaussian_color_stay_in_device
( sample, extrinsics, intrinsics, options={}, colors_overwrite=None, verbose=True,
sam3d_objects/model/backbone/tdfy_dit/utils/render_utils.py:124
Functionrender_snapshot
( samples, resolution=512, bg_color=(0, 0, 0), offset=(-16 / 180 * np.pi, 20 / 180 * np.pi),
sam3d_objects/model/backbone/tdfy_dit/utils/render_utils.py:199
Functionrender_video
( sample, resolution=512, bg_color=(0, 0, 0), num_frames=300, r=2.0, fov=40, pitch
notebook/inference.py:162
Functionrender_video
( sample, resolution=512, bg_color=(0, 0, 0), num_frames=300, r=2, fov=40, backend
sam3d_objects/model/backbone/tdfy_dit/utils/render_utils.py:155
Functionresize_all_to_same_size
Resize RGB image, mask, and pointmap to the same size. This is crucial when pointmaps have different resolution than RGB images, whi
sam3d_objects/data/dataset/tdfy/img_and_mask_transforms.py:431
Functionrev_lognorm_sampler
(mean=0.0, std=1.0, **kwargs)
sam3d_objects/model/backbone/generator/flow_matching/model.py:32
Methodrun
( self, image: Union[None, Image.Image, np.ndarray], mask: Union[None, Image.Image, np
sam3d_objects/pipeline/inference_pipeline_pointmap.py:385
Functionsafe_state
(silent)
sam3d_objects/model/backbone/tdfy_dit/representations/gaussian/general_utils.py:123
Functionscale_module
Scale the parameters of a module and return it.
sam3d_objects/model/backbone/tdfy_dit/modules/utils.py:62
Functionset_attn
(attn: Literal["xformers", "flash_attn"])
sam3d_objects/model/backbone/tdfy_dit/modules/sparse/__init__.py:55
Functionset_backend
(backend: Literal["xformers", "flash_attn", "torch_flash_attn"])
sam3d_objects/model/backbone/tdfy_dit/modules/attention/__init__.py:36
Functionset_backend
(backend: Literal["spconv", "torchsparse"])
sam3d_objects/model/backbone/tdfy_dit/modules/sparse/__init__.py:45
Functionset_debug
(debug: bool)
sam3d_objects/model/backbone/tdfy_dit/modules/attention/__init__.py:41
Functionset_debug
(debug: bool)
sam3d_objects/model/backbone/tdfy_dit/modules/sparse/__init__.py:50
Methodshape
(self)
sam3d_objects/model/backbone/tdfy_dit/modules/sparse/basic.py:155
Functionsimplify_gs
Simplify 3D Gaussians NOTE: this function is not used in the current implementation for the unsatisfactory performance. Args: gs
sam3d_objects/model/backbone/tdfy_dit/utils/postprocessing_utils.py:685
Functionsoftplus_inverse_scaling_activation
(x)
sam3d_objects/model/backbone/tdfy_dit/representations/gaussian/gaussian_model.py:250
Functionsparse_batch_op
Broadcast a 1D tensor to a sparse tensor along the batch dimension then perform an operation. Args: input (torch.Tensor): 1D tensor
sam3d_objects/model/backbone/tdfy_dit/modules/sparse/basic.py:472
Functionsparse_cat
Concatenate a list of sparse tensors. Args: inputs (List[SparseTensor]): List of sparse tensors to concatenate.
sam3d_objects/model/backbone/tdfy_dit/modules/sparse/basic.py:486
Methodssi_to_metric
(scale: torch.Tensor, shift: torch.Tensor)
sam3d_objects/data/dataset/tdfy/pose_target.py:473
Methodstate_dict
(self)
sam3d_objects/model/backbone/tdfy_dit/representations/octree/octree_dfs.py:250
Methodstep
(self, dynamics_fn, x_t, t, dt, *args, **kwargs)
sam3d_objects/model/backbone/generator/flow_matching/solver.py:48
Methodstep
(self, dynamics_fn, x_t, t, dt, *args, **kwargs)
sam3d_objects/model/backbone/generator/flow_matching/solver.py:87
Methodstep
(self, dynamics_fn, x_t, t, dt, *args, **kwargs)
sam3d_objects/model/backbone/generator/flow_matching/solver.py:115
Methodstructure_control
Control the structure of the octree. Args: mask (torch.Tensor): the mask to control the structure. 1 for subdivide, -1 f
sam3d_objects/model/backbone/tdfy_dit/representations/octree/octree_dfs.py:330
Functionsubkey
(mapping, key)
sam3d_objects/config/utils.py:144
Functionto_device
Recursively moves all tensors in obj to the specified device. Args: obj: Object to move to device - can be a tensor, list, tuple, dict or
sam3d_objects/data/utils.py:68
Functionto_glb
Convert a generated asset to a glb file. Args: app_rep (Union[Strivec, Gaussian]): Appearance representation. mesh (MeshExtr
sam3d_objects/model/backbone/tdfy_dit/utils/postprocessing_utils.py:585
Methodto_instance_pose
(invariant_targets: "InvariantPoseTarget")
sam3d_objects/data/dataset/tdfy/pose_target.py:240
Methodto_instance_pose
(cls, pose_target: PoseTarget)
sam3d_objects/data/dataset/tdfy/pose_target.py:278
Methodto_instance_pose
(cls, pose_target: PoseTarget, normalize: bool = False)
sam3d_objects/data/dataset/tdfy/pose_target.py:325
Methodto_instance_pose
(cls, pose_target: PoseTarget, normalize: bool = False)
sam3d_objects/data/dataset/tdfy/pose_target.py:423
Methodto_instance_pose
(cls, pose_target: PoseTarget, normalize: bool = False)
sam3d_objects/data/dataset/tdfy/pose_target.py:523
Methodto_invariant
(cls, instance_pose: InstancePose)
sam3d_objects/data/dataset/tdfy/pose_target.py:269
Methodto_invariant
(cls, pose_target: PoseTarget, normalize: bool = False)
sam3d_objects/data/dataset/tdfy/pose_target.py:351
Methodto_invariant
(cls, pose_target: PoseTarget)
sam3d_objects/data/dataset/tdfy/pose_target.py:449
Methodto_invariant
(cls, pose_target: PoseTarget, normalize: bool = False)
sam3d_objects/data/dataset/tdfy/pose_target.py:549
Methodto_invariant
(cls, pose_target: PoseTarget)
sam3d_objects/data/dataset/tdfy/pose_target.py:581
Methodto_invariant
(cls, pose_target: PoseTarget)
sam3d_objects/data/dataset/tdfy/pose_target.py:649
Methodto_invariant
(cls, pose_target: PoseTarget)
sam3d_objects/data/dataset/tdfy/pose_target.py:676
Methodto_invariant
(cls, pose_target: PoseTarget)
sam3d_objects/data/dataset/tdfy/pose_target.py:719
Functiontree_transpose_level_one
( structure, check_children=False, map_fn=None, is_leaf=None, )
sam3d_objects/data/utils.py:19
Functiontrimesh2o3d_mesh
(trimesh_mesh)
sam3d_objects/pipeline/inference_utils.py:772
Functionunnormalize_img
(img)
sam3d_objects/data/dataset/tdfy/img_and_mask_transforms.py:27
Functionupdate_layout
(pred_t, pred_s, pred_quat, center, scale, to_halo=True)
sam3d_objects/pipeline/inference_utils.py:780
Functionvisualize_meshes_comparison
Simple visualization of both meshes in a single 3D plot. DEPRECATED: Use visualize_meshes_interactive() for better interactive visualization
notebook/mesh_alignment.py:407
Functionvisualize_meshes_interactive
Interactive Gradio-based 3D visualization of aligned human and object meshes. Args: aligned_mesh_path: Path to aligned mesh PLY file
notebook/mesh_alignment.py:295
Functionvoxelize_mesh
(mesh, resolution=64)
sam3d_objects/pipeline/inference_utils.py:736
Methodwrapped
(*args, **kwargs)
sam3d_objects/pipeline/inference_pipeline.py:216
Functionzero_out
(args, kwargs)
sam3d_objects/model/backbone/generator/classifier_free_guidance.py:19
← previous701–784 of 784, ranked by callers