Functions — github.com/TencentARC/InstantMesh

hub / github.com/TencentARC/InstantMesh / functions

Functions275 in github.com/TencentARC/InstantMesh

⨍ Functions 275◇ Types & classes 49

↓ 10 callersMethod__init__

(self, config: ViTConfig)

src/models/encoder/dino.py:308

↓ 10 callersFunctioninstantiate_from_config

(config)

src/utils/train_util.py:11

↓ 6 callersFunctionFOV_to_intrinsics

Creates a 3x3 camera intrinsics matrix from the camera field of view, specified in degrees. Note the intrinsics are returned as normalized by

src/utils/camera_util.py:84

↓ 6 callersFunctionrank_zero_print

(*args)

train.py:17

↓ 5 callersMethod__init__

(self, state_dict: Dict[str, torch.Tensor])

zero123plus/pipeline.py:228

↓ 4 callersFunctionextract_into_tensor

(a, t, x_shape)

zero123plus/model.py:37

↓ 4 callersMethodforward_geometry

Main function of our Generator. It first generate 3D mesh, then render it into 2D image with given `render_cameras`. :param p

src/models/lrm_mesh.py:270

↓ 4 callersMethodload_im

replace background pixel with random color in rendering

src/data/objaverse.py:110

↓ 3 callersFunctioncenter_looking_at_camera_pose

Create OpenGL camera extrinsics from camera locations and look-at position. camera_position: (M, 3) or (3,) look_at: (3) up_world: (

src/utils/camera_util.py:16

↓ 3 callersMethodforward_planes

(self, images, cameras)

src/models/lrm.py:68

↓ 3 callersFunctionget_circular_camera_poses

(M=120, radius=2.5, elevation=30.0)

src/utils/camera_util.py:62

↓ 3 callersMethodget_texture_prediction

Predict Texture given triplanes :param planes: the triplane feature map :param tex_pos: Position we want to query the texture

src/models/lrm_mesh.py:192

↓ 3 callersMethodinit_flexicubes_geometry

(self, device, fovy=50.0)

src/models/lrm_mesh.py:77

↓ 3 callersFunctionsample_from_planes

(plane_axes, plane_features, coordinates, mode='bilinear', padding_mode='zeros', box_warp=None)

src/models/renderer/utils/renderer.py:62

↓ 3 callersFunctionsort_edges

(edges_ex2)

src/models/geometry/rep_3d/dmtet.py:47

↓ 3 callersMethodtranspose_for_scores

(self, x: torch.Tensor)

src/models/encoder/dino.py:172

↓ 2 callersMethod_forward_pass

Additional filtering is applied to filter out-of-box samples. Modifications made by Zexin He.

src/models/renderer/utils/renderer.py:121

↓ 2 callersMethod_linear_interp

Computes the location of zero-crossings on 'edges_x' using linear interpolation with 'edges_weight'.

src/models/geometry/rep_3d/flexicubes.py:331

↓ 2 callersMethodactivation_factory

(options: dict)

src/models/renderer/utils/renderer.py:114

↓ 2 callersMethodencode_condition_image

(self, image: torch.Tensor)

zero123plus/pipeline.py:322

↓ 2 callersMethodencode_condition_image

(self, images)

zero123plus/model.py:135

↓ 2 callersMethodextract_mesh

Extract a 3D mesh from triplane nerf. Only support batch_size 1. :param planes: triplane features :param mesh_resolution: mar

src/models/lrm.py:153

↓ 2 callersMethodforward

(self, lrm_generator_input)

src/model_mesh.py:185

↓ 2 callersMethodforward

(self, lrm_generator_input)

src/model.py:169

↓ 2 callersMethodforward_points

(self, planes, points: torch.Tensor, chunk_size: int = 2**20)

src/models/renderer/synthesizer.py:178

↓ 2 callersFunctiongenerate_planes

Defines planes by the three vectors that form the "axes" of the plane. Should work with arbitrary number of planes and planes of arbitrar

src/models/renderer/utils/renderer.py:28

↓ 2 callersMethodget_geometry_prediction

Function to generate mesh with give triplanes :param planes: triplane features

src/models/lrm_mesh.py:159

↓ 2 callersMethodget_mesh

(self, v_deformed_nx3, sdf_n, with_uv=False, indices=None)

src/models/geometry/rep_3d/dmtet.py:434

↓ 2 callersFunctionget_zero123plus_input_cameras

Get the input camera parameters.

src/utils/camera_util.py:95

↓ 2 callersFunctioninterpolate

(attr, rast, attr_idx, rast_db=None)

src/models/geometry/render/neural_render.py:17

↓ 2 callersFunctionmodulate

(x, shift, scale)

src/models/encoder/dino.py:301

↓ 2 callersMethodprepare

(self)

zero123plus/pipeline.py:312

↓ 2 callersMethodprepare_batch_data

(self, batch)

zero123plus/model.py:103

↓ 2 callersFunctionremove_background

(image: PIL.Image.Image, rembg_session: Any = None, force: bool = False, **rembg_kwargs, )

src/utils/infer_util.py:11

↓ 2 callersFunctionresize_foreground

( image: PIL.Image.Image, ratio: float, )

src/utils/infer_util.py:25

↓ 2 callersMethodrun_forward

(self, colors, densities, depths, rendering_options, normals=None)

src/models/renderer/utils/ray_marcher.py:30

↓ 2 callersMethodrun_model

(self, planes, decoder, sample_coordinates, sample_directions, options)

src/models/renderer/utils/renderer.py:197

↓ 2 callersMethodsample_stratified

Return depths of approximately uniformly spaced samples along rays.

src/models/renderer/utils/renderer.py:239

↓ 2 callersFunctionsave_obj

(pointnp_px3, facenp_fx3, colornp_px3, fpath)

src/utils/mesh_util.py:18

↓ 2 callersFunctionto_rgb_image

(maybe_rgba: Image.Image)

zero123plus/pipeline.py:30

↓ 2 callersFunctionunscale_image

(image)

zero123plus/model.py:32

↓ 2 callersFunctionxatlas_uvmap

(ctx, mesh_v, mesh_pos_idx, resolution)

src/models/geometry/rep_3d/extract_texture_map.py:20

↓ 1 callersMethod__getattr__

(self, name: str)

zero123plus/pipeline.py:190

↓ 1 callersMethod__init__

(self, savedir)

train.py:114

↓ 1 callersMethod__init__

(self, root_dir='objaverse/', meta_fname='valid_paths.json', image_dir='rendering_zero

src/data/objaverse_zero123plus.py:61

↓ 1 callersMethod__init__

(self, root_dir='objaverse/', meta_fname='valid_paths.json', input_image_dir='renderin

src/data/objaverse.py:75

↓ 1 callersMethod__init__

(self, n_features: int, hidden_dim: int = 64, num_layers: int = 4, activation: nn.Module = nn

src/models/renderer/synthesizer_mesh.py:23

↓ 1 callersMethod__init__

(self, n_features: int, hidden_dim: int = 64, num_layers: int = 4, activation: nn.Module = nn

src/models/renderer/synthesizer.py:25

↓ 1 callersMethod__init__

(self)

src/models/renderer/utils/ray_sampler.py:23

↓ 1 callersMethod__init__

( self, inner_dim: int, image_feat_dim: int, triplane_low_res: int,

src/models/decoder/transformer.py:73

↓ 1 callersMethod_build_activation_factory

(self)

src/models/renderer/utils/renderer.py:113

↓ 1 callersMethod_build_dino

(model_name: str, proxy_error_retries: int = 3, proxy_error_cooldown: int = 5)

src/models/encoder/dino_wrapper.py:67

↓ 1 callersMethod_compute_reg_loss

Regularizer L_dev as in Equation 8

src/models/geometry/rep_3d/flexicubes.py:218

↓ 1 callersMethod_compute_vd

Computes the location of dual vertices as described in Section 4.2

src/models/geometry/rep_3d/flexicubes.py:357

↓ 1 callersMethod_freeze

(self)

src/models/encoder/dino_wrapper.py:60

↓ 1 callersMethod_get_case_id

Obtains the ID of topology cases based on cell corner occupancy. This function resolves the ambiguity in the Dual Marching Cubes (DM

src/models/geometry/rep_3d/flexicubes.py:252

↓ 1 callersMethod_identify_surf_cubes

Identifies grid cubes that intersect with the underlying surface by checking if the signs at all corners are not identical.

src/models/geometry/rep_3d/flexicubes.py:320

↓ 1 callersMethod_identify_surf_edges

Identifies grid edges that intersect with the underlying surface by checking for opposite signs. As each edge can be shared by multi

src/models/geometry/rep_3d/flexicubes.py:295

↓ 1 callersMethod_normalize_weights

Normalizes the given weights to be non-negative. If input weights are None, it creates and returns a set of weights of ones.

src/models/geometry/rep_3d/flexicubes.py:228

↓ 1 callersMethod_solve_vd_QEF

(self, p_bxnx3, norm_bxnx3, c_bx3=None)

src/models/geometry/rep_3d/flexicubes.py:343

↓ 1 callersMethod_tetrahedralize

Tetrahedralizes the interior volume to produce a tetrahedral mesh, as described in Section 4.5.

src/models/geometry/rep_3d/flexicubes.py:500

↓ 1 callersMethod_triangulate

Connects four neighboring dual vertices to form a quadrilateral. The quadrilaterals are then split into triangles based on the gamma

src/models/geometry/rep_3d/flexicubes.py:449

↓ 1 callersMethodcompute_loss

(self, noise_pred, noise_gt)

zero123plus/model.py:226

↓ 1 callersMethodcompute_loss

(self, render_out, render_gt)

src/model_mesh.py:243

↓ 1 callersMethodcompute_loss

(self, render_out, render_gt)

src/model.py:243

↓ 1 callersFunctioncompute_vertex_normal

(v_pos, t_pos_idx)

src/models/geometry/render/neural_render.py:42

↓ 1 callersMethodconstruct_voxel_grid

Generates a voxel grid based on the specified resolution. Args: res (int or list[int]): The resolution of the voxel grid

src/models/geometry/rep_3d/flexicubes.py:100

↓ 1 callersFunctioncreate_mt_variable

(device)

src/models/geometry/rep_3d/dmtet.py:20

↓ 1 callersFunctioncreate_tetmesh_variables

(device='cuda')

src/models/geometry/rep_3d/dmtet.py:107

↓ 1 callersFunctiondot

(x, y)

src/models/geometry/render/neural_render.py:38

↓ 1 callersMethodencode_target_images

(self, images)

zero123plus/model.py:144

↓ 1 callersMethodforward_cond

(self, noisy_cond_lat, timestep, encoder_hidden_states, class_labels, ref_dict, is_cfg_guidance, **kwargs)

zero123plus/pipeline.py:106

↓ 1 callersMethodforward_grid

(self, planes, grid_size: int, aabb: torch.Tensor = None)

src/models/renderer/synthesizer.py:146

↓ 1 callersMethodforward_lrm_generator

(self, images, cameras, render_cameras, render_size=512)

src/model_mesh.py:171

↓ 1 callersMethodforward_lrm_generator

( self, images, cameras, render_cameras, render_size=192,

src/model.py:136

↓ 1 callersMethodforward_planes

(self, images, cameras)

src/models/lrm_mesh.py:88

↓ 1 callersMethodforward_synthesizer

(self, planes, render_cameras, render_size: int)

src/models/lrm.py:82

↓ 1 callersMethodforward_unet

(self, latents, t, prompt_embeds, cond_latents)

zero123plus/model.py:153

↓ 1 callersMethodforward_vision_encoder

(self, images)

zero123plus/model.py:120

↓ 1 callersFunctionget_center_boundary_index

(verts)

src/models/geometry/rep_3d/dmtet_utils.py:12

↓ 1 callersFunctionget_center_boundary_index

(grid_res, device)

src/models/geometry/rep_3d/flexicubes_geometry.py:17

↓ 1 callersMethodget_file_list

(self)

train.py:117

↓ 1 callersMethodget_geometry_prediction

(self, sampled_features, flexicubes_indices)

src/models/renderer/synthesizer_mesh.py:69

↓ 1 callersMethodget_mesh

(self, v_deformed_nx3, sdf_n, weight_n=None, with_uv=False, indices=None, is_training=False)

src/models/geometry/rep_3d/flexicubes_geometry.py:63

↓ 1 callersFunctionget_obj_from_str

(string, reload=False)

src/utils/train_util.py:21

↓ 1 callersFunctionget_parser

(**parser_kwargs)

train.py:21

↓ 1 callersFunctionget_render_cameras

Get the rendering camera parameters.

run.py:25

↓ 1 callersFunctionget_render_cameras

Get the rendering camera parameters.

app.py:38

↓ 1 callersMethodget_sdf_deformation_prediction

Predict SDF and deformation for tetrahedron vertices :param planes: triplane feature map for the geometry

src/models/lrm_mesh.py:102

↓ 1 callersFunctionget_tet_bxfx4x3

(bxnxz, bxfx4)

src/models/geometry/rep_3d/dmtet.py:328

↓ 1 callersMethodget_texture_prediction

Predict Texture given triplanes :param planes: the triplane feature map :param tex_pos: Position we want to query the texture

src/models/lrm.py:107

↓ 1 callersMethodget_texture_prediction

(self, sampled_features)

src/models/renderer/synthesizer_mesh.py:83

↓ 1 callersMethodget_v

(self, x, noise, t)

zero123plus/model.py:174

↓ 1 callersFunctionimages_to_video

( images: torch.Tensor, output_path: str, fps: int = 30, )

src/utils/infer_util.py:66

↓ 1 callersFunctioninterpolate

(attr, rast, attr_idx, rast_db=None)

src/utils/mesh_util.py:157

↓ 1 callersFunctioninterpolate

(attr, rast, attr_idx, rast_db=None)

src/models/geometry/rep_3d/extract_texture_map.py:16

↓ 1 callersMethodinterpolate_pos_encoding

This method allows to interpolate the pre-trained position encodings, to be able to use the model on higher resolution images.

src/models/encoder/dino.py:50

↓ 1 callersMethodload_im

(self, path, color)

src/data/objaverse_zero123plus.py:87

↓ 1 callersMethodload_im

replace background pixel with random color in rendering

src/data/objaverse.py:280

↓ 1 callersFunctionmake_mesh

(mesh_fpath, planes)

app.py:155

next →1–100 of 275, ranked by callers