![\"\"](\"https:\/\/3dcoat.com\/documentation\/wp-content\/uploads\/2024\/09\/Photogrammetry-Menu.jpg\")
<\/figure>\n<\/div>\n\n\nTo start ensure that all the necessary conditions for Photogrammetry to work correctly are met.<\/p>\n\n\n
![\"\"](\"https:\/\/3dcoat.com\/documentation\/wp-content\/uploads\/2024\/09\/image-2.png\")
<\/figure>\n<\/div>\n\n\nPhotogrammetry Left Tool Panel<\/h3>\n\n\n\n
\n\n\n\n![\"\"](\"https:\/\/3dcoat.com\/documentation\/wp-content\/uploads\/2024\/09\/Photo-Left-Panel.jpg\")
<\/figure>\n<\/div>\n\n\n\n- Transform: <\/strong>Transform the object as an entire. Click on the object to transform. Choose a rectangle or lasso mode to select multiple objects at once and transform them. Use SHIFT, CTRL CRTL+SHIFT to add, subtract or intersect selection.<\/li>\n<\/ul>\n\n\n\n
\n- Photogrammetry tool: <\/strong>\u0410utomatic Image-based 3D reconstruction tool from images.<\/li>\n<\/ul>\n\n\n\n
3DC use external library of software for photogrammetry processing: Image-based 3D reconstruction are based on Colmap<\/a><\/strong> and Reality Capture<\/a><\/strong>.<\/p>\n\n\n\n![\"\"](\"https:\/\/3dcoat.com\/documentation\/wp-content\/uploads\/2024\/09\/Photogrammetry-tool.jpg\")
<\/figure>\n<\/div>\n\n\nReality Capture engine<\/h3>\n\n\n\n
\n\n\n\n\n- Video to 3D: Extract sharp frames from a video and create a 3D model from them.
3D Coat, unlike Reality Capture, selects the sharpest frames rather than just any at the same time interval, but this process should take several minutes. <\/li>\n\n\n\n - Shots to 3D: Make a 3D model from photos. Once the process is finished a popup panel ask to increase or not the texture resolution for better reconstruction.<\/li>\n<\/ul>\n\n\n\n
![\"\"](\"https:\/\/3dcoat.com\/documentation\/wp-content\/uploads\/2024\/09\/import-the-mesh-with-texture.png\")
<\/figure>\n<\/div>\n\n\n
– Import Project: Create a new object linked to an existing photogrammetry project.
– Bake UV Textures: Bake texture from source photos into PPP model.
The PPP model must match the transform with the associated photogrammetry project.<\/p>\n\n\n\n
Colmap engine<\/h3>\n\n\n\n
\n\n\n\n
\u0410utomatic reconstruction tool that simply takes a folder of input images and produces a sparse and dense reconstruction in a workspace folder.
The output is written to the workspace folder.
For example, if your images are located in path\/to\/project\/images<\/em>, you could select path\/to\/project<\/em> as a workspace folder.<\/p>\n\n\n\nPoint Cloud tool<\/h3>\n\n\n\n
\n\n\n\n
3DCoat have built-in point cloud-to-mesh conversion tools under development on Photogrammetry Room.<\/p>\n\n\n\n
Point Cloud tool:<\/strong> Use this tool to create meshes from the point clouds and convert the LAS point cloud to 3D TIN file using Poisson reconstruction algorithms.<\/p>\n\n\n\n![\"\"](\"https:\/\/3dcoat.com\/documentation\/wp-content\/uploads\/2024\/09\/image-3.png?wsr\")
<\/figure>\n<\/div>\n\n\nPoint cloud file name:<\/strong> Select the point cloud file to begin the conversion.
Poisson depth:<\/strong> This integer is the maximum depth of the tree that will be used for surface reconstruction.
Poisson trim:<\/strong> This floating point values specifies the value for mesh trimming.<\/p>\n\n\n\n
\n\n\n\n“Poisson depth”<\/strong> is a key parameter in the Poisson surface reconstruction<\/strong> algorithm. It refers to the octree depth<\/strong> used to discretize the 3D space and store the reconstructed surface mesh<\/mark>. <\/p>\n\n\n\n\n- Function:<\/strong> The Poisson reconstruction method uses an adaptive octree to partition the space and solve the Poisson equation, which determines an implicit indicator function (a function that is 1 inside the object and 0 outside).<\/li>\n\n\n\n
- Resolution:<\/strong> The
depth<\/code> parameter controls the resolution of the resulting 3D mesh. A higher octree depth creates a finer, more detailed surface mesh but requires exponentially more memory and computation time.<\/li>\n\n\n\n- Impact:<\/strong>\n
\n- Low depth:<\/strong> Results in a coarse, less detailed mesh.<\/li>\n\n\n\n
- High depth:<\/strong> Results in a high-resolution, detailed mesh, but increases computational load and memory usage.<\/li>\n<\/ul>\n<\/li>\n\n\n\n
- Practical Use:<\/strong> Users typically adjust the
depth<\/code> value to balance the desired quality of the reconstructed surface with available computational resource<\/li>\n<\/ul>\n\n\n\n
\n\n\n\n“Poisson depth trim” refers to the process of using a density threshold to trim away “excess” surfaces or artifacts in regions of low point cloud density after the initial reconstruction process<\/strong>. <\/p>\n\n\n\nExplanation<\/p>\n\n\n\n
\n- Poisson Surface Reconstruction:<\/strong> The Poisson algorithm works by creating a smooth, watertight<\/em> (completely closed) 3D mesh from an oriented point cloud. This process can create surfaces that fill in actual holes in the data, but it can also generate unwanted floating artifacts or extraneous surfaces in areas where the original point cloud data was sparse or missing.<\/li>\n\n\n\n
- Density Information:<\/strong> During the Poisson reconstruction, a density value is calculated for each vertex in the resulting mesh. This value indicates the level of support from the original input points for that specific part of the reconstructed surface. A low density value means the vertex is in an area with few or no nearby input points.<\/li>\n\n\n\n
- Trimming:<\/strong> The “trim” part of the process involves a post-processing step where a user-defined threshold is applied to these density values. Any vertices, and their connected triangles, with a density value below this specified threshold are removed from the mesh.<\/li>\n\n\n\n
- Result:<\/strong> This effectively “cuts away” the unwanted, low-support surfaces while preserving the main object, which is supported by a high density of points. <\/li>\n\n\n\n
- <\/li>\n<\/ul>\n\n\n\n
The term “depth” is a reference to the octree depth<\/strong> parameter used in the Poisson algorithm itself, which controls the maximum resolution of the reconstruction grid. While related to the reconstruction quality, the “depth trim” specifically relates to using the density output<\/em> from the reconstruction to trim the final mesh.<\/p>\n\n\n\n
\n\n\n\nAt current stage of development it is not as effective as solution found on other apps.<\/p>\n\n\n\n
To create a 3D model from a point cloud in 3DCoat, you will first need to use another application to generate the mesh from the point cloud.<\/p>\n\n\n\n
Once you have a mesh file (e.g., .obj, .fbx, .ply), you can import it into 3DCoat. In 3DCoat, you can then clean up the mesh, retopologize it to create a lower-polygon model with cleaner topology, and then use 3DCoat’s tools for sculpting, texturing, and painting on the new mesh.<\/p>\n\n\n\n
Step 1<\/strong>: Generate a mesh from the point cloud in another program<\/p>\n\n\n\nPreprocess the point cloud: Use a program like CloudCompare, MeshLab, or others to clean up the point cloud by removing noise and non-essential points.<\/p>\n\n\n\n
Create the mesh: Generate a mesh from the cleaned point cloud using the program’s built-in tools. Some programs have algorithms like Poisson reconstruction or surface reconstruction that can create a surface from the points.
Export the mesh: Export the generated mesh as a standard 3D file format like .obj or .fbx.<\/p>\n\n\n\n
This video demonstrates how to create a 3D mesh from point clouds in another application:<\/p>\n\n\n\n
<\/figure>\n<\/div>\n\n\n- \n
- Transform: <\/strong>Transform the object as an entire. Click on the object to transform. Choose a rectangle or lasso mode to select multiple objects at once and transform them. Use SHIFT, CTRL CRTL+SHIFT to add, subtract or intersect selection.<\/li>\n<\/ul>\n\n\n\n
- \n
- Photogrammetry tool: <\/strong>\u0410utomatic Image-based 3D reconstruction tool from images.<\/li>\n<\/ul>\n\n\n\n
3DC use external library of software for photogrammetry processing: Image-based 3D reconstruction are based on Colmap<\/a><\/strong> and Reality Capture<\/a><\/strong>.<\/p>\n\n\n
\n<\/figure>\n<\/div>\n\n\nReality Capture engine<\/h3>\n\n\n\n
\n\n\n\n- \n
- Video to 3D: Extract sharp frames from a video and create a 3D model from them.
3D Coat, unlike Reality Capture, selects the sharpest frames rather than just any at the same time interval, but this process should take several minutes. <\/li>\n\n\n\n - Shots to 3D: Make a 3D model from photos. Once the process is finished a popup panel ask to increase or not the texture resolution for better reconstruction.<\/li>\n<\/ul>\n\n\n\n<\/figure>\n<\/div>\n\n\n
– Import Project: Create a new object linked to an existing photogrammetry project.
– Bake UV Textures: Bake texture from source photos into PPP model.
The PPP model must match the transform with the associated photogrammetry project.<\/p>\n\n\n\nColmap engine<\/h3>\n\n\n\n
\n\n\n\n\u0410utomatic reconstruction tool that simply takes a folder of input images and produces a sparse and dense reconstruction in a workspace folder.
The output is written to the workspace folder.
For example, if your images are located in path\/to\/project\/images<\/em>, you could select path\/to\/project<\/em> as a workspace folder.<\/p>\n\n\n\nPoint Cloud tool<\/h3>\n\n\n\n
\n\n\n\n3DCoat have built-in point cloud-to-mesh conversion tools under development on Photogrammetry Room.<\/p>\n\n\n\n
Point Cloud tool:<\/strong> Use this tool to create meshes from the point clouds and convert the LAS point cloud to 3D TIN file using Poisson reconstruction algorithms.<\/p>\n\n\n
\n<\/figure>\n<\/div>\n\n\nPoint cloud file name:<\/strong> Select the point cloud file to begin the conversion.
Poisson depth:<\/strong> This integer is the maximum depth of the tree that will be used for surface reconstruction.
Poisson trim:<\/strong> This floating point values specifies the value for mesh trimming.<\/p>\n\n\n\n
\n\n\n\n“Poisson depth”<\/strong> is a key parameter in the Poisson surface reconstruction<\/strong> algorithm. It refers to the octree depth<\/strong> used to discretize the 3D space and store the reconstructed surface mesh<\/mark>. <\/p>\n\n\n\n
- \n
- Function:<\/strong> The Poisson reconstruction method uses an adaptive octree to partition the space and solve the Poisson equation, which determines an implicit indicator function (a function that is 1 inside the object and 0 outside).<\/li>\n\n\n\n
- Resolution:<\/strong> The
depth<\/code> parameter controls the resolution of the resulting 3D mesh. A higher octree depth creates a finer, more detailed surface mesh but requires exponentially more memory and computation time.<\/li>\n\n\n\n- Impact:<\/strong>\n
- \n
- Low depth:<\/strong> Results in a coarse, less detailed mesh.<\/li>\n\n\n\n
- High depth:<\/strong> Results in a high-resolution, detailed mesh, but increases computational load and memory usage.<\/li>\n<\/ul>\n<\/li>\n\n\n\n
- Practical Use:<\/strong> Users typically adjust the
depth<\/code> value to balance the desired quality of the reconstructed surface with available computational resource<\/li>\n<\/ul>\n\n\n\n
\n\n\n\n“Poisson depth trim” refers to the process of using a density threshold to trim away “excess” surfaces or artifacts in regions of low point cloud density after the initial reconstruction process<\/strong>. <\/p>\n\n\n\n
Explanation<\/p>\n\n\n\n
- \n
- Poisson Surface Reconstruction:<\/strong> The Poisson algorithm works by creating a smooth, watertight<\/em> (completely closed) 3D mesh from an oriented point cloud. This process can create surfaces that fill in actual holes in the data, but it can also generate unwanted floating artifacts or extraneous surfaces in areas where the original point cloud data was sparse or missing.<\/li>\n\n\n\n
- Density Information:<\/strong> During the Poisson reconstruction, a density value is calculated for each vertex in the resulting mesh. This value indicates the level of support from the original input points for that specific part of the reconstructed surface. A low density value means the vertex is in an area with few or no nearby input points.<\/li>\n\n\n\n
- Trimming:<\/strong> The “trim” part of the process involves a post-processing step where a user-defined threshold is applied to these density values. Any vertices, and their connected triangles, with a density value below this specified threshold are removed from the mesh.<\/li>\n\n\n\n
- Result:<\/strong> This effectively “cuts away” the unwanted, low-support surfaces while preserving the main object, which is supported by a high density of points. <\/li>\n\n\n\n
- <\/li>\n<\/ul>\n\n\n\n
The term “depth” is a reference to the octree depth<\/strong> parameter used in the Poisson algorithm itself, which controls the maximum resolution of the reconstruction grid. While related to the reconstruction quality, the “depth trim” specifically relates to using the density output<\/em> from the reconstruction to trim the final mesh.<\/p>\n\n\n\n
\n\n\n\nAt current stage of development it is not as effective as solution found on other apps.<\/p>\n\n\n\n
To create a 3D model from a point cloud in 3DCoat, you will first need to use another application to generate the mesh from the point cloud.<\/p>\n\n\n\n
Once you have a mesh file (e.g., .obj, .fbx, .ply), you can import it into 3DCoat. In 3DCoat, you can then clean up the mesh, retopologize it to create a lower-polygon model with cleaner topology, and then use 3DCoat’s tools for sculpting, texturing, and painting on the new mesh.<\/p>\n\n\n\n
Step 1<\/strong>: Generate a mesh from the point cloud in another program<\/p>\n\n\n\n
Preprocess the point cloud: Use a program like CloudCompare, MeshLab, or others to clean up the point cloud by removing noise and non-essential points.<\/p>\n\n\n\n
Create the mesh: Generate a mesh from the cleaned point cloud using the program’s built-in tools. Some programs have algorithms like Poisson reconstruction or surface reconstruction that can create a surface from the points.
Export the mesh: Export the generated mesh as a standard 3D file format like .obj or .fbx.<\/p>\n\n\n\nThis video demonstrates how to create a 3D mesh from point clouds in another application:<\/p>\n\n\n\n
- Density Information:<\/strong> During the Poisson reconstruction, a density value is calculated for each vertex in the resulting mesh. This value indicates the level of support from the original input points for that specific part of the reconstructed surface. A low density value means the vertex is in an area with few or no nearby input points.<\/li>\n\n\n\n
- High depth:<\/strong> Results in a high-resolution, detailed mesh, but increases computational load and memory usage.<\/li>\n<\/ul>\n<\/li>\n\n\n\n
- Resolution:<\/strong> The
- Function:<\/strong> The Poisson reconstruction method uses an adaptive octree to partition the space and solve the Poisson equation, which determines an implicit indicator function (a function that is 1 inside the object and 0 outside).<\/li>\n\n\n\n
- Video to 3D: Extract sharp frames from a video and create a 3D model from them.
- Photogrammetry tool: <\/strong>\u0410utomatic Image-based 3D reconstruction tool from images.<\/li>\n<\/ul>\n\n\n\n