应用所有数据增强

应用图像模糊

应用轮廓增强

应用图像填充

应用径向畸变

应用图像旋转

Compute grid size, bounding box minimum coordinates, and bounding box size based on input bounds and resolution. Args:

根据方位角计算旋转结果

连接到材质输出节点

创建基础色混合shader

创建金属度粗糙度组合RGB节点

创建完整的MR和BC节点结构

提取mask并合成到白色背景

查找已存在的自定义节点

查找BSDF和输出节点

获取或创建MR和BC节点

获取渲染路径和版本

获取所有数据选择和增强标志

检查图像是否在过滤字典中

处理所有选中的图像

处理单个材质

处理输出节点链接

过滤并选择图像

设置场景视图

判断是否应该跳过该图像

Shuffle the data in the stream. This uses a buffer of size `bufsize`. Shuffling at startup is less random; this is traded off against

Split off all file extensions. Returns base, allext. Args: path: path with extensions Returns: path with all extensions

Translate the vertices so that bounding box is centered at zero.

Compute Chamfer distance between predicted mesh and ground-truth mesh. pred_vertices/pred_faces: mesh A (prediction) gt_vertices/gt_face

清理分布式环境

Compute the ratio between d_alpha and alpha

Compute the diffusion term of the SDE Args: x: [batch_dim, ...], data point t: [batch_dim,], time vector form: s

compute ulip and uni3d score for a list of image and mesh paths image_paths: list of image paths mesh_paths: list of mesh pat

Compute the mean of time-dependent density p_t

compute ulip score between point cloud and images pc: (1, N, 3) torch tensor images: list of PIL images

compute uni3d score between point cloud and images pc: (1, N, 3) torch tensor images: list of PIL images

Compute the vector field corresponding to p_t

Sample xt from time-dependent density p_t; rng is required

Converts all objects in the scene to meshes. Returns: None

Deletes all invisible objects in the scene. Returns: None

r""" Offloads all models to CPU using accelerate, reducing memory usage with a low impact on performance. Compared to `enable_sequenti

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

member function for obtaining the drift of the probability flow ODE

See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298 Args: timeste

member function for obtaining score of x_t = alpha_t * x + sigma_t * eps

Wrapper function: transfrom velocity prediction model to score Args: velocity: [batch_dim, ...] shaped tensor; velocity model outp

Group tarfile contents by keys and yield samples. Args: data: iterator over tarfile contents keys: function that takes a file nam

Guess the batch size by looking at the length of the first element in a tuple.

Resets the scene to a clean state. Returns: None

加载模型（单卡推理，不进行模型并行）

Loads a model with a supported file extension into the scene. Args: object_path (str): Path to the model file. Raises: Value

主函数：加载渲染图像

Normalize the mesh, sample points and normals from its surface, and randomly select a subset. Args: mesh (trimesh.Trimesh): Input me