Meshopt simplify

The first simplification algorithm, meshopt_simplify, follows the topology of the original mesh in an attempt to preserve attribute seams, borders and overall appearance. For meshes with inconsistent topology or many seams, such as faceted meshes, it can result in simplifier getting "stuck" and not being able to simplify the mesh fully. Therefore it's critical that identical vertices are "welded" together, that is, the input vertex buffer does not contain duplicates. Additionally, it may be possible to preprocess the index buffer (e.g. with meshopt_generateShadowIndexBuffer) to discard any vertex attributes that aren't critical and can be rebuilt later.

第一个简化算法meshopt_simplify遵循原始网格的拓扑结构，试图保留属性接缝、边界和整体外观。

• 对于拓扑不一致或有许多缝隙(seams)的网格，可能会导致算法卡死(stuck)，并且无法完成简化
• 因此，建议对索引缓冲区进行预处理meshopt_generateShadowIndexBuffer，以丢弃不重要的、再之后还可以重新的顶点属性（如法线）。

float threshold = 0.2f;
size_t target_index_count = size_t(index_count * threshold);
float target_error = 1e-2f;

std::vector<unsigned int> lod(index_count);
float lod_error = 0.f;
lod.resize(meshopt_simplify(&lod[0], indices, index_count, &vertices[0].x, vertex_count, sizeof(Vertex),
    target_index_count, target_error, &lod_error));

参数

https://github.com/zeux/meshoptimizer/pull/216 This change adds an extra optional argument, , to that can be used to output the error after the simplification, which - just like target_error - is a measure of the relative (0..1) deviation of the produced result from the source mesh. The argument is optional in C++ but required when C interface is used - NULL can be passed if this information isn't required.result_errormeshopt_simplify When absolute errors are required instead, newly added can be used to scale the input or the output error back into world space.meshopt_simplifyScale

https://github.com/zeux/meshoptimizer/issues/59 The error that the simplifier computes is somewhat hard to define exactly because it's based on a normalized quadric error. When just a single edge collapse is involved, it's just a square of the distance, but when multiple collapses are involved it becomes something like a quadratic mean I think. I'm fuzzy about this because the normalization of the quadric is unusal and doesn't occur in the literature, and I haven't derived what exactly happens after a series of collapses :) It's great to know that it's close enough though - it can definitely be useful to determine the LOD switching distances. I will look into exposing this properly.

target_error

Target error is an approximate measure of the deviation from the original mesh using distance normalized to [0..1] range (e.g. 1e-2f means that simplifier will try to maintain the error to be below 1% of the mesh extents). Note that the simplifier attempts to produce the requested number of indices at minimal error, but because of topological restrictions and error limit it is not guaranteed to reach the target index count and can stop earlier.

对原始网格偏差的近似度量归一化到[0,1]中，故target_error的阈值是[0,1] target_error=1e-2f，表示简化器会尽量将误差保持在网格范围的1%以下

v0.11

/**
* Experimental: Mesh simplifier
* Reduces the number of triangles in the mesh, attempting to preserve mesh appearance as much as possible
    减少网格中的三角形数量，尽量保持网格外观
* The algorithm tries to preserve mesh topology and can stop short of the target goal based on topology constraints or target error.
    尽量保持网络的拓扑结构，并根据拓扑约束或目标误差来组织目标目的的不同
* If not all attributes from the input mesh are required, it's recommended to reindex the mesh using meshopt_generateShadowIndexBuffer prior to simplification.
    如果输入网格的所有属性不是必须的，建议在简化之前使用meshopt_generateShadowIndexBuffer重新索引网格。
* Returns the number of indices after simplification, with destination containing new index data
    返回简化的索引数，目标包含新的索引数据
* The resulting index buffer references vertices from the original vertex buffer.
    产生的索引缓冲区引用来自原始顶点缓冲区的顶点
* If the original vertex data isn't required, creating a compact vertex buffer using meshopt_optimizeVertexFetch is recommended.
    如果不需要原始顶点数据，建议使用meshopt_optimizeVertexFetch创建一个紧凑的顶点缓冲区
*
* destination must contain enough space for the *source* index buffer (since optimization is iterative, this means index_count elements - *not* target_index_count!)
    destination必须包含足够的空间用于*source* index缓冲区(因为优化是迭代的，这意味着index_count元素- *不是* target_index_count!)
* vertex_positions should have float3 position in the first 12 bytes of each vertex - similar to glVertexPointer
    vertex_positions应该有float3位置在每个顶点的前12个字节-类似于glVertexPointer
*/
MESHOPTIMIZER_EXPERIMENTAL size_t meshopt_simplify(
    unsigned int* destination,      //简化后的三角网索引，存在哪
    const unsigned int* indices,    //原始三角网索引
    size_t index_count,             //原始索引个数
    const float* vertex_positions,  //原始顶点坐标
    size_t vertex_count,            //原始顶点个数
    size_t vertex_positions_stride, //顶点对齐大小
    size_t target_index_count,      //目标索引个数
    float target_error,             //目标误差
    bool * topologyres=nullptr
);

v0.18

v0.18版本以后，新增了lod_error参数

float threshold = 0.2f;
size_t target_index_count = size_t(index_count * threshold);
float target_error = 1e-2f;

std::vector<unsigned int> lod(index_count);
float lod_error = 0.f;
lod.resize(meshopt_simplify(&lod[0], indices, index_count, &vertices[0].x, vertex_count, sizeof(Vertex),
    target_index_count, target_error, &lod_error));