MCPcopy Create free account

hub / github.com/TencentARC/Pixal3D / functions

Functions865 in github.com/TencentARC/Pixal3D

⨍ Functions 865◇ Types & classes 147↳ Endpoints 3
↓ 3 callersMethodget_proj_cond_shape
Get proj conditioning for shape/texture stages (sparse-token aligned). Args: image_cond_model: The proj image cond model
pixal3d/pipelines/pixal3d_image_to_3d.py:230
↓ 3 callersFunctionget_renderer
(sample, **kwargs)
pixal3d/utils/render_utils.py:43
↓ 3 callersFunctionindent
Indent a string.
pixal3d/utils/general_utils.py:365
↓ 3 callersFunctionlpips
(img1, img2, value_range=(0, 1))
pixal3d/utils/loss_utils.py:73
↓ 3 callersFunctionmodel_grads_to_master_grads
Copy the gradients from the model parameters into the master parameters from make_master_params().
pixal3d/trainers/utils.py:59
↓ 3 callersFunctionparse_view_indices
Parse view_indices string into a sorted deduplicated list of integers.
data_toolkit/utils.py:433
↓ 3 callersMethodpreprocess_image
Preprocess the input image. Args: input: Input image (RGB or RGBA). bg_color: Background color (R, G, B) in
pixal3d/pipelines/pixal3d_image_to_3d.py:144
↓ 3 callersMethodreduce
(self, op: str, dim: Optional[Union[int, Tuple[int,...]]] = None, keepdim: bool = False)
pixal3d/modules/sparse/basic.py:267
↓ 3 callersMethodrender
Render the gausssian. Args: voxel (Voxel): Voxel representation. extrinsics (torch.Tensor): (4, 4) camera ex
pixal3d/renderers/voxel_renderer.py:24
↓ 3 callersFunctionsparse_scaled_dot_product_attention
Apply scaled dot product attention to a sparse tensor. Args: qkv (VarLenTensor): A [N, *, 3, H, C] sparse tensor containing Qs, Ks,
pixal3d/modules/sparse/attention/full_attn.py:13
↓ 3 callersFunctionsparse_windowed_scaled_dot_product_self_attention
Apply windowed scaled dot product self attention to a sparse tensor. Args: qkv (SparseTensor): [N, *, 3, H, C] sparse tensor contain
pixal3d/modules/sparse/attention/windowed_attn.py:74
↓ 3 callersFunctionssim
(img1, img2, window_size=11, size_average=True)
pixal3d/utils/loss_utils.py:39
↓ 3 callersMethodupsample
(self, x: sp.SparseTensor, upsample_times: int)
pixal3d/models/sc_vaes/sparse_unet_vae.py:509
↓ 3 callersMethodvisualize_sample
Convert a sample to an image.
pixal3d/trainers/basic.py:564
↓ 2 callersMethod__cal_layout
(self, coords, batch_size)
pixal3d/modules/sparse/basic.py:467
↓ 2 callersMethod__cal_shape
(self, feats, coords)
pixal3d/modules/sparse/basic.py:461
↓ 2 callersMethod__init__
(self, channels: int, mlp_ratio: float = 4.0)
pixal3d/modules/sparse/transformer/blocks.py:12
↓ 2 callersMethod__init__
(self, roots: str, *, resolution: int, min_aesthetic_score: float = 5.0,
pixal3d/datasets/structured_latent_shape.py:185
↓ 2 callersMethod__init__
(self, roots: str, *, min_aesthetic_score: float = 5.0, normalization: Optiona
pixal3d/datasets/sparse_structure_latent.py:232
↓ 2 callersFunction_finish_progress
()
app.py:297
↓ 2 callersMethod_forward_with_mem_ratio
Forward with a given memory ratio.
pixal3d/utils/elastic_utils.py:167
↓ 2 callersMethod_fused_pre
(self, x: Union[VarLenTensor, torch.Tensor], num_fused: int)
pixal3d/modules/sparse/attention/modules.py:91
↓ 2 callersMethod_get_feats
(self, t)
pixal3d/modules/sparse/attention/proj_attention.py:76
↓ 2 callersMethod_master_params_to_state_dicts
Convert master params to dict of state_dicts.
pixal3d/trainers/basic.py:314
↓ 2 callersMethod_pred_to_xstart
(self, x_t, t, pred)
pixal3d/pipelines/samplers/flow_euler.py:38
↓ 2 callersFunction_progress_file
Return path to a session's progress JSON file.
app.py:284
↓ 2 callersMethod_randomize_camera
(self, num_samples: int)
pixal3d/trainers/vae/shape_vae.py:113
↓ 2 callersMethod_randomize_camera
(self, num_samples: int)
pixal3d/trainers/vae/pbr_vae.py:103
↓ 2 callersMethod_remap_checkpoint_keys
Remap checkpoint keys to match model state dict. Handles structural changes like: - cross_attn.xxx -> cross_attn.cro
pixal3d/trainers/basic.py:446
↓ 2 callersFunction_reset_progress
(session_id: str)
app.py:288
↓ 2 callersMethod_reshape_chs
(x: Union[VarLenTensor, torch.Tensor], shape: Tuple[int, ...])
pixal3d/modules/sparse/attention/modules.py:85
↓ 2 callersMethod_rotary_embedding
(self, x: torch.Tensor, phases: torch.Tensor)
pixal3d/modules/sparse/attention/rope.py:29
↓ 2 callersMethod_updown
(self, x: sp.SparseTensor, subdiv: sp.SparseTensor = None)
pixal3d/models/sc_vaes/sparse_unet_vae.py:59
↓ 2 callersMethodapply_rotary_embedding
(x: torch.Tensor, phases: torch.Tensor)
pixal3d/modules/attention/rope.py:29
↓ 2 callersFunctionclear_cuda_error
()
data_toolkit/encode_shape_latent.py:23
↓ 2 callersFunctionclear_cuda_error
()
data_toolkit/encode_shape_latent_view.py:26
↓ 2 callersFunctionclear_cuda_error
()
data_toolkit/encode_ss_latent.py:21
↓ 2 callersFunctionclear_cuda_error
()
data_toolkit/encode_pbr_latent.py:23
↓ 2 callersFunctionclear_cuda_error
()
data_toolkit/encode_ss_latent_view.py:20
↓ 2 callersFunctionclear_cuda_error
()
data_toolkit/encode_pbr_latent_view.py:26
↓ 2 callersFunctioncompute_text_block
Wrap the entire text (splitting at explicit newline characters) using the provided scale, and then compute the overall width and heig
data_toolkit/utils.py:343
↓ 2 callersFunctioncompute_text_block
Wrap the entire text (splitting at explicit newline characters) using the provided scale, and then compute the overall width and heig
pixal3d/utils/general_utils.py:248
↓ 2 callersMethodcpu
(self)
pixal3d/modules/sparse/basic.py:145
↓ 2 callersMethodcuda
(self)
pixal3d/modules/sparse/basic.py:149
↓ 2 callersMethoddevice
(self)
pixal3d/trainers/basic.py:214
↓ 2 callersFunctiondistance_from_fov
(camera_angle_x, grid_point, target_point, mesh_scale, image_resolution)
app.py:213
↓ 2 callersFunctiondistance_from_fov
(camera_angle_x, grid_point, target_point, mesh_scale, image_resolution)
inference.py:124
↓ 2 callersMethodencode_image
Encode the image.
pixal3d/trainers/flow_matching/mixins/image_conditioned.py:144
↓ 2 callersMethodencode_image_proj
Encode the image with view-aligned projection. Supports both 'proj' mode (DINOv3 only, 2 outputs) and 'gated_proj'
pixal3d/trainers/flow_matching/mixins/image_conditioned_proj.py:1276
↓ 2 callersMethodencode_images
Encode the image.
pixal3d/trainers/flow_matching/mixins/image_conditioned.py:198
↓ 2 callersFunctionfind_ckpt
(cfg)
train.py:24
↓ 2 callersMethodforward
(self, x: torch.Tensor)
pixal3d/modules/norm.py:7
↓ 2 callersMethodforward
(self, input: VarLenTensor)
pixal3d/modules/sparse/nonlinearity.py:14
↓ 2 callersMethodforward
(self, *args, **kwargs)
pixal3d/utils/elastic_utils.py:176
↓ 2 callersMethodget_cond
Get the conditioning information for the model. Args: image (Union[torch.Tensor, list[Image.Image]]): The image prompts.
pixal3d/pipelines/trellis2_image_to_3d.py:175
↓ 2 callersMethodget_cond
Get the conditioning information for the model. Args: image (Union[torch.Tensor, list[Image.Image]]): The image prompts.
pixal3d/pipelines/trellis2_texturing.py:159
↓ 2 callersMethodget_mem_ratio
(self, input_size)
pixal3d/utils/elastic_utils.py:29
↓ 2 callersMethodhalf
(self)
pixal3d/modules/sparse/basic.py:153
↓ 2 callersFunctionis_power_of_two
(n: int)
pixal3d/datasets/sparse_voxel_pbr.py:19
↓ 2 callersFunctionis_valid_sparse_tensor
(tensor)
data_toolkit/encode_shape_latent.py:20
↓ 2 callersFunctionis_valid_sparse_tensor
(tensor)
data_toolkit/encode_shape_latent_view.py:23
↓ 2 callersFunctionload_latent
Load a latent .npz file and return a SparseTensor on GPU.
data_toolkit/visualize_pbr_latent.py:41
↓ 2 callersFunctionload_moge_model
(device="cuda", model_name=MOGE_MODEL_NAME)
app.py:113
↓ 2 callersFunctionmain
(local_rank, cfg)
train.py:61
↓ 2 callersFunctionmaster_params_to_model_params
Copy the master parameter data back into the model parameters.
pixal3d/trainers/utils.py:49
↓ 2 callersFunctionproject_points_to_image_batch
Project 3D points to 2D image coordinates (batch processing). Args: points_3d: torch.Tensor, shape [N, 3] or [B, N, 3], 3D point
pixal3d/trainers/flow_matching/mixins/image_conditioned_proj.py:27
↓ 2 callersMethodread_pbr_voxel
(self, root, instance)
pixal3d/datasets/sparse_voxel_pbr.py:255
↓ 2 callersMethodrender
Render the mesh. Args: mesh : meshmodel extrinsics (torch.Tensor): (4, 4) camera extrinsics intr
pixal3d/renderers/pbr_mesh_renderer.py:233
↓ 2 callersMethodrun_snapshot
Run a snapshot of the model.
pixal3d/trainers/basic.py:557
↓ 2 callersMethodrun_step
Run a training step.
pixal3d/trainers/basic.py:992
↓ 2 callersMethodsample
Generate samples from the model using Euler method. Args: model: The model to sample from. noise: Th
pixal3d/pipelines/samplers/flow_euler.py:84
↓ 2 callersMethodsample_shape_slat
Sample structured latent with the given conditioning. Args: cond (dict): The conditioning information.
pixal3d/pipelines/trellis2_image_to_3d.py:248
↓ 2 callersMethodsample_shape_slat_cascade
Sample structured latent with the given conditioning. Args: cond (dict): The conditioning information.
pixal3d/pipelines/trellis2_image_to_3d.py:288
↓ 2 callersMethodsave_logs
(self)
pixal3d/trainers/basic.py:1118
↓ 2 callersFunctionscene_bbox
Returns the bounding box of the scene. Taken from Shap-E rendering script (https://github.com/openai/shap-e/blob/main/shap_e/rendering/blende
data_toolkit/blender_script/dump_pbr.py:115
↓ 2 callersFunctionscene_bbox
Returns the bounding box of the scene. Taken from Shap-E rendering script (https://github.com/openai/shap-e/blob/main/shap_e/rendering/blende
data_toolkit/blender_script/dump_mesh.py:114
↓ 2 callersFunctionscene_bbox
Returns the bounding box of the scene. Taken from Shap-E rendering script (https://github.com/openai/shap-e/blob/main/shap_e/rendering/blende
data_toolkit/blender_script/render_cond.py:290
↓ 2 callersFunctionsphere_normalize_torch
Sphere normalization: normalize vertices based on sphere radius.
data_toolkit/utils.py:542
↓ 2 callersMethodstd
(self, dim: Optional[Union[int, Tuple[int,...]]] = None, keepdim: bool = False)
pixal3d/modules/sparse/basic.py:295
↓ 2 callersFunctionstr_to_dtype
(dtype_str: str)
pixal3d/modules/utils.py:77
↓ 2 callersMethodto
(self, device)
pixal3d/modules/image_feature_extractor.py:22
↓ 2 callersMethodupdate_run_states
(self, input_size=None, mem_ratio=None)
pixal3d/utils/elastic_utils.py:21
↓ 2 callersMethodvisualize_projection
Visualize the projected 3D grid points on the input image. Args: image: Input image tensor [B, C, H, W], assumed
pixal3d/trainers/flow_matching/mixins/image_conditioned_proj.py:234
↓ 2 callersMethodvisualize_projection_test
Visualize projection points on the condition images. This should be called once before training starts to verify the project
pixal3d/trainers/flow_matching/mixins/image_conditioned_proj.py:1481
↓ 1 callersMethod__cal_spatial_shape
(self, coords)
pixal3d/modules/sparse/basic.py:473
↓ 1 callersFunction__from_env
()
pixal3d/modules/attention/config.py:6
↓ 1 callersFunction__from_env
()
pixal3d/modules/sparse/config.py:7
↓ 1 callersFunction__getattr__
(name)
pixal3d/models/__init__.py:24
↓ 1 callersMethod__init__
(self, cross_attn_block: nn.Module, channels: int, proj_in_channels: int)
pixal3d/modules/attention/proj_attention.py:33
↓ 1 callersMethod__init__
(self, dim: int, heads: int)
pixal3d/modules/attention/modules.py:10
↓ 1 callersMethod__init__
(self, num_groups, num_channels, eps=1e-5, affine=True)
pixal3d/modules/sparse/norm.py:16
↓ 1 callersMethod__init__
(self, cross_attn_block: nn.Module, channels: int, proj_in_channels: int)
pixal3d/modules/sparse/attention/proj_attention.py:21
↓ 1 callersMethod__init__
(self, dim: int, heads: int)
pixal3d/modules/sparse/attention/modules.py:12
↓ 1 callersMethod__init__
(self, factor: int = 2)
pixal3d/modules/sparse/spatial/spatial2channel.py:12
↓ 1 callersMethod__init__
( self, factor: int )
pixal3d/modules/sparse/spatial/basic.py:76
↓ 1 callersMethod__init__
(self, in_channels, out_channels, kernel_size, stride=1, dilation=1, padding=None, bias=True, indice_key=None)
pixal3d/modules/sparse/conv/conv.py:12
↓ 1 callersMethod__init__
( self, channels: int, num_heads: int, mlp_ratio: float = 4.0, attn_mo
pixal3d/modules/sparse/transformer/modulated.py:14
↓ 1 callersMethod__init__
( self, channels: int, num_heads: int, mlp_ratio: float = 4.0, attn_mo
pixal3d/modules/transformer/modulated.py:13
↓ 1 callersMethod__init__
(self)
pixal3d/utils/elastic_utils.py:152
↓ 1 callersMethod__init__
( self, dataset: Dataset, shuffle: bool = True, seed: int = 0, drop_la
pixal3d/utils/data_utils.py:83
← previousnext →101–200 of 865, ranked by callers