Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/uhlik/bpy
blender python scripts
https://github.com/uhlik/bpy
addon blender blender-addon blender-scripts opengl photogrammetry ply-files point-cloud point-cloud-visualizer time-tracker uv-mapping wavefront-obj zbrush
Last synced: 3 months ago
JSON representation
blender python scripts
- Host: GitHub
- URL: https://github.com/uhlik/bpy
- Owner: uhlik
- Created: 2014-06-12T04:11:05.000Z (over 10 years ago)
- Default Branch: master
- Last Pushed: 2023-07-04T22:23:57.000Z (over 1 year ago)
- Last Synced: 2024-01-26T03:03:22.519Z (9 months ago)
- Topics: addon, blender, blender-addon, blender-scripts, opengl, photogrammetry, ply-files, point-cloud, point-cloud-visualizer, time-tracker, uv-mapping, wavefront-obj, zbrush
- Language: Python
- Size: 36.8 MB
- Stars: 907
- Watchers: 43
- Forks: 104
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome - uhlik/bpy - blender python scripts (Python)
- awesome - uhlik/bpy - blender python scripts (Python)
README
## [Point Cloud Visualizer 2.0+](https://www.blendermarket.com/products/pcv)
**Supported, maintained and updated to work in latest Blender versions. Many, many more features.**
Available here: [https://www.blendermarket.com/products/pcv](https://www.blendermarket.com/products/pcv)
Full documentation is here: [https://jakubuhlik.com/docs/pcv/docs.html](https://jakubuhlik.com/docs/pcv/docs.html)
***
***
# **All addons available in this repository are unmaintained and unsupported, some may work in recent Blender versions some may not, some are obsolete.**
***
***
# Addons for blender 2.80 and 2.81 (maybe up to 2.83)
* [Point Cloud Visualizer](#point-cloud-visualizer-for-blender-280)
* [Color Management Presets](#color-management-presets-for-blender-280)
* [Tube UV Unwrap](#tube-uv-unwrap-for-blender-280)
* [Fast Wavefront^2](#fast-wavefront2-for-blender-280)
* [Import Agisoft PhotoScan Cameras](#import-agisoft-photoscan-cameras)
* [Carbon Tools](#carbon-tools)
* [Time Tracker](#time-tracker-for-blender-280)
***
***
## [Point Cloud Visualizer 0.9.x](https://raw.githubusercontent.com/uhlik/bpy/master/space_view3d_point_cloud_visualizer.py) (for blender 2.80)
**Display, edit, filter, render, convert, generate and export colored point cloud PLY files.**
Works with any PLY file with 'x, y, z, nx, ny, nz, red, green, blue' vertex values. Vertex normals and colors are optional.
# This is old, unmaintained and unsupported version
# This version fully works only in **Blender 2.81**
### General info
##### Basic Usage:
* Install and activate addon in a usual way.
* Add any object type to scene.
* Go to 3d View Sidebar (N) > `Point Cloud Visualizer` tab, click file browser icon, select ply file, click `Load PLY`. `Reload` button next to it reloads ply from disk.
* Click `Draw` button to display point cloud, `Erase` to hide point cloud. Adjust percentage of displayed points with `Display`, point size with `Size` and point transparency with `Alpha`.
* Display point normals as lines - click `Normal` icon, adjust line length with `Length` next to it.
* Transforming parent object transforms point cloud as well.
* `Illumination` 'adds' single artificial light on points, you can edit its direction and strength. Works only when vertex normals are present.
* When vertex colors are missing, cloud will be displayed in uniform gray, in that case you can enable `Illumination` to have better cloud view
##### Display Options:
* `Display` - percentage of displayed points
* `Size` - point size in pixels
* `Alpha` - global points alpha
* `Normals` - display point normals as lines, adjust line length with `Length` next to it
* `Illumination` - enable extra illumination, works only when vertex normals can be loaded
* `Light Direction` - illumination light direction
* `Light Intensity` - illumination light intensity
* `Shadow Intensity` - illumination shadow intensity
* `Depth` - enable depth debug shader
* `Brightness` - depth shader color brightness
* `Contrast` - depth shader color contrast
* `False Colors` - display depth shader in false colors
* `Color A` - depth shader false colors front color
* `Color B` - depth shader false colors back color
* `Normal` - enable normal debug shader
* `Position` - enable position debug shader
### Shaders
### Clip
Shader with 6 clipping planes. Planes can be enabled/disabled independently with eye icon, first 3 values are plane normal (x, y, z), last value is plane distance from origin. These can be set from different object bounding box, choose object from scene in `Object` and hit `Set Clip Planes From Object Bounding Box`, object can then be hidden/deleted, set values will stay until operator is executed again. `X` reset all the settings to their defaults.
* `Enable Clipping Planes Shader` - enable shader
* `Plane 0` - plane 0 normal (x, y, z), plane 0 distance from origin
* `Plane 1` - plane 1 normal (x, y, z), plane 1 distance from origin
* `Plane 2` - plane 2 normal (x, y, z), plane 2 distance from origin
* `Plane 3` - plane 3 normal (x, y, z), plane 3 distance from origin
* `Plane 4` - plane 4 normal (x, y, z), plane 4 distance from origin
* `Plane 5` - plane 5 normal (x, y, z), plane 5 distance from origin
* `Object` - object to use as bounding box source
* `Set Clip Planes From Object Bounding Box` - set planes from selected object bounding box
* `X` - reset all the settings to their defaults
### Edit
Quasi point cloud Edit Mode. Hit `Start` and all points are converted to helper mesh with vertices and entered to mesh edit mode. You can transform, delete and duplicate vertices using regular Blender's tools. If you want update displayed points, hit `Update`, when you are finished editing hit `End` to update points for a last time and delete helper mesh. If something went wrong, select main object with cloud and hit `Cancel` to reload original points, return interface to regular mode and attempt to clean helper mesh if it is still available.
* To save edited point cloud, you have to use `Export` feature and check `Use Viewport Points` because edits are only in memory, if you close Blender, edits will be lost.
* Point normals are not changed (at this time), if you rotate points, normals will be still oriented as before.
* New points can be reliably (for now) created by duplicating existing points. If you create new points, they will all have the same random normal and random color.
`Start` - Start edit mode, create helper object and switch to it
`Overlay Alpha` - Overlay point alpha
`Overlay Size` - Overlay point size
`Update` - Update displayed cloud from edited mesh
`End` - Update displayed cloud from edited mesh, stop edit mode and remove helper object
`Cancel` - Stop edit mode, try to remove helper object and reload original point cloud
### Filter
Filter current point cloud, all changes are only temporary, original data are still intact. To keep changes, you have to export cloud as ply file.
##### Simplify
Simplify point cloud to exact number of evenly distributed samples. All loaded points are processed. Higher samples counts may take a long time to finish. Surely there are better (and faster) tools for the job, but.. Basically it takes some random point and set as accepted sample, then another set of random candidates, measure distance from already accepted samples and stores the one that is most distant as another accepted, repeat until number of samples is reached.
* `Samples` - Number of points in simplified point cloud, best result when set to less than 20% of points, when samples has value close to total expect less points in result
* `Candidates` - Number of candidates used during resampling, the higher value, the slower calculation, but more even
* `Simplify` - run operator
##### Project
Project points on mesh (or object convertible to mesh) surface. Projects point along their normals until it hit surface or `Search Distance` is reached. You can choose between `Positive` (along normal direction), `Negative` (vice versa) or both. Optionally you can `Discard Unprojectable` points that was not possible to project and after projection `Shift` points a fixed distance along normal (positive value) or the other way around (negative value). Projected points can be optionally colorized by vertex colors, uv texture and vertex group from target mesh.
* `Object` - Mesh or object convertible to mesh
* `Search Distance` - Maximum search distance in which to search for surface
* `Positive` - Search along point normal forwards
* `Negative` - Search along point normal backwards
* `Discard Unprojectable` - Discard points which didn't hit anything
* `Colorize` - Colorize projected points
* `Source` - Color source for projected point cloud
- `Vertex Colors` - Use active vertex colors from target
- `UV Texture` - Use colors from active image texture node in active material using active UV layout from target
- `Vertex Group Monochromatic` - Use active vertex group from target, result will be shades of grey
- `Vertex Group Colorized` - Use active vertex group from target, result will be colored from red (1.0) to blue (0.0) like in weight paint viewport
* `Shift` - Shift points after projection above (positive value) or below (negative value) surface
* `Project` - execute operator
##### Boolean
Intersect or Exclude points with mesh object. Mesh have to be non-manifold.
* `Object` - Mesh or object convertible to mesh
* `Intersect` - Keep points inside mesh, remove points outside
* `Exclude` - Keep points outside mesh, remove points inside
##### Remove Color
Remove points with exact/similar color as chosen in color picker (Eyedropper works too). Currently i can't get to match sampled color from viewport with color in loaded cloud. Floating point error, incorrectly handled Gamma (at my side for sure), color management in Blender's viewport or any combination of all, or something else.. Anyway, if you leave at least one delta at hue/saturation/value (whatever works best for given cloud) it should remove the color you picked.
* `Color` - Color to remove from point cloud
* `Δ Hue` - Delta hue
* `Δ Saturation` - Delta saturation
* `Δ Value` - Delta value
* `Select Color` - Run operator and display selected points in viewport for review
* `X` - Deselect
* `Delete Selected` - Delete selected points
##### Merge
Load another ply and merge with currently displayed. Hit `Merge With Other PLY`, select ply file and load. New point will be appended to old, shuffled if shuffle is enabled in preferences.
* `Merge With Other PLY` - run operator
##### Join
Join current active point cloud with another on different object. Object transformations are applied to clouds to look the same as before joining.
* `Object` - select object with another point cloud to join
* `Join` - run operator
##### Color Adjustment
Adjust exposure, gamma, brightness, contrast, hue, saturation, value or invert colors. Click `Enable` to enable color adjustment shader, adjust values as needed, click `Apply` to apply colors to points. Click `Reset` to set all to default value. Shader can be disabled without changes anytime unchecking `Enable`. When color adjustment shader is enabled, all other shaders are disabled.
* `Enabled` - Enable color adjustment shader, other shaders will be overrided until disabled
* `Exposure` - formula: color = color * (2 ** value)
* `Gamma` - formula: color = color ** (1 / value)
* `Brightness` - formula: color = (color - 0.5) * contrast + 0.5 + brightness
* `Contrast` - formula: color = (color - 0.5) * contrast + 0.5 + brightness
* `Hue` - formula: color.h = (color.h + (value % 1.0)) % 1.0
* `Saturation` - formula: color.s += value
* `Value` - formula: color.v += value
* `Invert` - formula: color = 1.0 - color
* `Reset` - Reset color adjustment values
* `Apply` - Apply color adjustments to points, reset and exit
### Render
Currently only sigle point cloud per render/frame is supported. Output image is RGBA 8bit PNG - transparent background with colored point cloud, which can be composed over something else later.
##### Usage:
* Blend file has to be saved
* Load and display ply first.
* Make a camera and adjust as needed.
* Select cloud parent object, set point size with `Size` or percentage of rendered points with `Count`
* Set render path in `Output`.
* Set render image size with `Resolution X`, `Resolution Y` and `Resolution %`.
* At default resolution settings are taken from scene, to make them independent, click chain icon next to properties, correct aspect ratio is not calculated, if you link properties again, values are copied from scene.
* If `Illumination` is enabled it will be rendered as well
* If one of debug shaders is enabled, it will be rendered instead of regular shader
* Hit `Render` or `Animation`
##### Render options:
* `Count` - percentage of rendered points
* `Size` - point render size in pixels
* `Supersampling` - Render larger image and then resize back to anti-alias, 1 - disabled, 2 - render 200%, 3 - render 300%, etc.
* `Smooth Circles` - Currently works only for basic shader with/without illumination and generally is much slower than Supersampling, use only when Supersampling fails
* `Output` - path where to save rendered images, `#` characters defines the position and length of frame numbers, image is always saved, filetype is always png, accepts relative paths, upon hitting `Render` path is validated, changed to absolute and written back
* `Resolution X` - image width in pixels
* `Resolution Y` - image height in pixels
* `Resolution %` - percentage scale for resolution
* `Resolution Linked` - when enabled, settings are taken from scene, if not they are independent on scene, but aspect ratio is not calculated
##### Render smoothing/anti-aliasing notes:
Use `Supersampling` for most of the time with lowest value (to be exact, `2` works great) that look still good. If you run into problems due to very large supersampled image sizes Blender or hardware cannot handle, you can try to disable `Supersampling` and use `Smooth Circles` to draw smooth points directly in target resolution, but you are limited to basic shader and points before drawing have to be depth sorted manually which can be very slow with large point counts.
### Convert
Convert point cloud to mesh. May result in very large meshes, e.g. 1m point cloud to cubes = 8m poly mesh. Depending on what point cloud data is available and desired mesh type, some options may not be enabled.
Conversion to instancer specifics: points are converted to triangle mesh object, vertex colors are baked to texture, extra instanced sphere object is added as child object of main mesh, material using baked colors is added to sphere and each instance inherits color of corresponding face it is instanced from.
Conversion to particles specifics: points are converted to triangle mesh object, vertex colors are baked to texture, particle system is added to mesh with one particle on each face, extra instanced sphere added as child object of main mesh and particle system is set to render that sphere, material using baked colors is added to sphere and each instance inherits color of corresponding face it emit from. Result is regular particle system which can be further edited, e.g. instance mesh changed, physics added etc.
* `Type` - Instance mesh type, Vertex, Equilateral Triangle, Tetrahedron, Cube or Ico Sphere
* `All`, `Subset` - Use all points or random subset of by given percentage
* `Size` - Mesh instance size, internal instanced mesh has size 1.0 so if you set size to 0.01, resulting instances will have actual size of 0.01 event when cloud is scaled
* `Align To Normal` - Align instance to point normal, e.g. tetrahedron point will align to normal, triangle plane will align to normal etc.
* `Colors` - Assign point color to instance vertex colors, each instance will be colored by point color (except vertices)
* `Sphere Subdivisions` - Conversion to instancer / particles only, number of subdivisions of particle system instanced ico sphere
### Generate
Generate point cloud from mesh (or object convertible to mesh). To store point cloud, use `Export` to save as ply file.
* `Source` - Points generation source
- `Vertices` - Use mesh vertices
- `Surface` - Use triangulated mesh surface
- `Particles` - Use active particle system
* `Particles` `(Source: Particles)` - Particles source selection
- `All` `(Source: Particles)` - Use all particles
- `Alive` `(Source: Particles)` - Use only alive particles
* `Algorithm` `(Source: Surface)` - Point generating algorithm
- `Weighted Random In Triangle` - Average triangle areas to approximate number of random points in each to get even distribution of points. If some very small polygons are left without points, increase number of samples. Mesh is triangulated before processing, on non-planar polygons, points will not be exactly on original polygon surface.
- `Poisson Disk Sampling` - Warning: slow, very slow indeed.. Uses Weighted Random In Triangle algorithm to pregenerate samples with all its inconveniences.
* `Approximate Number Of Points` `(Source: Surface)` - Number of points to generate, some algorithms may not generate exact number of points.
* `Seed` `(Source: Surface)` - Random number generator seed
* `Colors` `(Source: Vertices, Surface, Particles)` - Color source for generated point cloud
- `Constant Color` - Use constant color value
- `Vertex Colors` - Use active vertex colors
- `UV Texture` - Generate colors from active image texture node in active material using active UV layout
- `Vertex Group Monochromatic` - Use active vertex group, result will be shades of grey
- `Vertex Group Colorized` - Use active vertex group, result will be colored from red (1.0) to blue (0.0) like in weight paint viewport
* `Color` `(Source: Vertices, Surface, Particles)` - Constant color
* `Exact Number of Samples` `(Source: Surface, Algorithm: Triangle)` - Generate exact number of points, if selected algorithm result is less points, more points will be calculated on random polygons at the end, if result is more points, points will be shuffled and sliced to match exact value
* `Minimal Distance` `(Source: Surface, Algorithm: Poisson)` - Poisson Disk minimal distance between points, the smaller value, the slower calculation
* `Sampling Exponent` `(Source: Surface, Algorithm: Poisson)` - Poisson Disk presampling exponent, lower values are faster but less even, higher values are slower exponentially
When `Source` is `Particles`, for generating colors (apart from `Constant` color), **non-overlapping UV layout** is required (can be really bad, useless for real production).
### Export
Export current point cloud as binary ply file with several options. If exporting modified (filtered) points, check `Use Viewport Points`, otherwise you will not get modified points. If exporting viewport points colors may slightly differ. Transformation and axis conversion can be applied on both loaded and viewport points.
* `Use Viewport Points` - When checked, export points currently displayed in viewport or when unchecked, export data loaded from original ply file
* `Visible Points Only` - Export currently visible points only (controlled by 'Display' on main panel)
* `Apply Transformation` - Apply parent object transformation to points
* `Convert Axes` - Convert from blender (y forward, z up) to forward -z, up y axes
### Sequence
Load sequence of ply files to play in viewport. Load first frame as regular file and when `Preload Sequence` is clicked it tries to load all ply files matching selected ply filename, e.g. you select `sequence-001.ply` and all `sequence-###.ply` will be loaded from directory. Only last number in filename is considered. Numbers should start at 1. All other features works when animation is not playing, but all changes are lost when you change frame to another.
* `Preload Sequence` - Load all matching ply files
* `Cycle Forever` - Cycle frames if timeline is longer than number of loaded frames
* `Clear Sequence` - Clear all loaded and return object to regular state i.e. you can load another ply, changes are kept etc.
### External API
To display point cloud data from other addons/custom scripts.
```python
import bpy
import numpy as np
from space_view3d_point_cloud_visualizer import PCVControl
o = bpy.context.active_object
c = PCVControl(o)
n = 100
vs = np.random.normal(0, 2, (n, 3))
ns = np.array([[0.0, 0.0, 1.0]] * n)
cs = np.random.random((n, 3))
# draw points
c.draw(vs, ns, cs)
```
```python
# if some data like normals/colors are not available
c.draw(vs, None, None)
c.draw(vs, [], [])
c.draw(vs)
# it is also possible to pass nothing in which case nothing is drawn
c.draw()
```
```python
# to stop any drawing
c.erase()
# to return object control to user
c.reset()
```
### Addon Preferences:
* `Default` - Default color to be used upon loading PLY to cache when vertex colors are missing
* `Normal` - Display color for vertex normals lines
* `Selection` - Display color for selection
* `Shuffle Points` - Shuffle points upon loading, display percentage is more useable if points are shuffled, disabled if you plan to export ply and you need to keep point order
* `Convert 16bit Colors` - Convert 16bit colors to 8bit, applied when Red channel has 'uint16' dtype
* `Gamma Correct 16bit Colors` - When 16bit colors are encountered apply gamma as 'c ** (1 / 2.2)'
* `Tab Name` - To have PCV in its own separate tab, choose one
* `Custom Tab Name` - Check if you want to have PCV in custom named tab or in existing tab
* `Name` - Custom PCV tab name, if you choose one from already existing tabs it will append to that tab
### Changelog:
* 0.9.30 join filter, ui tweaks, fixes
* 0.9.29 fixes, render logging
* 0.9.28 clipping planes from object bounding box, faster all numpy export, many minor fixes
* 0.9.27 render supersampling and draw smooth circles
* 0.9.26 color adjustment fixes
* 0.9.25 color adjustment, faster boolean
* 0.9.24 project colors
* 0.9.23 fixes
* 0.9.22 extra debug shaders
* 0.9.21 preferences
* 0.9.20 ui changes, poisson disk sampling generation, size and alpha available in edit mode, various fixes, removed manual depth sorting during render
* 0.9.19 point cloud generation from vertices and particles
* 0.9.18 point cloud generation from mesh surface
* 0.9.17 fixes
* 0.9.16 boolean filters
* 0.9.15 selection preview for remove color filter
* 0.9.14 external api improvements
* 0.9.13 faster normals drawing
* 0.9.12 ply sequence, external api
* 0.9.11 merge filter
* 0.9.10 ui
* 0.9.9 point cloud global alpha
* 0.9.8 basic editing
* 0.9.7 project point cloud on mesh surface
* 0.9.6 ply exporting now uses original or viewport data
* 0.9.5 simplify and remove color filters
* 0.9.4 export ply
* 0.9.3 conversion to instancer
* 0.9.2 load ply with 16bit colors
* 0.9.1 all new render settings
* 0.9.0 conversion to particles
* 0.8.14 fixes
* 0.8.13 fixes
* 0.8.12 fixes
* 0.8.11 ui tweaks
* 0.8.10 fixes
* 0.8.9 ui tweaks, code cleanup
* 0.8.8 refactored convert to mesh
* 0.8.7 fixed vcols bug in convert
* 0.8.6 ui tweaks, a few minor optimizations
* 0.8.5 convert to mesh all or subset
* 0.8.4 preferences, ui tweaks
* 0.8.3 display normals
* 0.8.2 fixed shader unknown attribute name
* 0.8.1 fixed ply with alpha, fixed convert to mesh when normals or colors are missing
* 0.8.0 convert to mesh
* 0.7.2 ui tweaks
* 0.7.1 viewport performance fixes
* 0.7.0 ascii ply support
* 0.6.6 fixed drawing after undo/redo
* 0.6.5 point cloud illumination
* 0.6.4 refactored draw handlers, fixed occasional crash on erase
* 0.6.3 added percentage of rendered points, fixed render colors to look the same as in viewport
* 0.6.2 fixed point size display in viewport, separated view and render point size
* 0.6.1 single cloud rendering almost completely rewritten to be better and faster
* 0.6.0 single cloud rendering
* 0.5.2 refactored some logic, removed icons from buttons
* 0.5.1 load ply without vertex colors, uniform grey will be used
* 0.5.0 performance improvements using numpy for loading and processing data
* 0.4.6 fixed crash when parent object is deleted while drawing, fixed removal of loaded data when parent is deleted
* 0.4.5 added 'Display' percentage, better error handling during .ply loading
* 0.4.0 almost complete rewrite for blender 2.80, performance improvements using shaders, simplified ui
* 0.3.0 new ply loader, can be used with any binary ply file with vertex coordinates and colors
* 0.2.0 display percentage
* 0.1.0 first release
##### Known bugs:
* If you duplicate object with cloud, duplicate will still control the original one until you load a different one. Currently there is no reliable way (as far as i know) to get unique id of an object and therefore no way to tell to which object stored properties (e.g. path to ply) belong.
[BlenderArtist.org thread](https://blenderartists.org/forum/showthread.php?416158-Addon-Point-Cloud-Visualizer)
***
## [Color Management Presets](https://raw.githubusercontent.com/uhlik/bpy/master/color_management_presets.py) (for blender 2.80)
Presets support for Render > Color Management panel, nothing more, nothing less.. Comes with a few presets i use which are created upon activation.
***
## [Tube UV Unwrap](https://raw.githubusercontent.com/uhlik/bpy/master/uv_tube_unwrap.py) (for blender 2.80)
UV unwrap tube-like meshes (all quads, no caps, fixed number of vertices in each ring)
notes:
* Works only on tube-like parts of mesh defined by selection and active vertex (therefore you must be in vertex selection mode) and the selection must have a start and an end ring. Tube-like mesh is: all quads, no caps, fixed number of vertices in each ring. (Best example of such mesh is mesh circle extruded several times or beveled curve (not cyclic) converted to mesh.) There must be an active vertex on one of the boundary loops in selection. This active vertex define place where mesh will be 'cut' - where seam will be placed.
* Result is rectangular UV for easy texturing, scaled to fit square, horizontal and vertical distances between vertices are averaged and proportional to each other.
usage:
1. tab to Edit mode
2. select part of mesh you want to unwrap, tube type explained above
3. make sure your selection has boundaries and there is an active vertex on one border of selection
4. hit "U" and select "Tube UV Unwrap"
5. optionally check/uncheck 'Mark Seams' or 'Flip' in operator properties
changelog:
* 0.3.0 blender 2.8 update
[BlenderArtist.org thread](http://blenderartists.org/forum/showthread.php?339782-UV-Equalize-and-Tube-Unwrap-addons)
***
## [Fast Wavefront^2](https://github.com/uhlik/bpy/tree/master/io_mesh_fast_obj) (for blender 2.80)
**Import/Export single mesh as Wavefront OBJ. Fast. Now with Cython. Binaries not included.**
Only active mesh is exported. Only single mesh is expected on import. Supported obj features: UVs, normals, vertex colors using MRGB format (ZBrush) or 'Extended' format (import only) where vertex is defined as (x,y,z,r,g,b).
changelog:
* 0.3.6 fixed export with applied modifiers
* 0.3.5 api changes fixes
* 0.3.4 ui
* 0.3.3 fallback python export implementation in case cython module is not available
* 0.3.2 import 'extended' vertex colors (x,y,z,r,g,b), optionally apply gamma correction
* 0.3.1 import obj (python only)
* 0.3.0 export implemented in cython
* 0.2.0 ported to blender 2.80
**installation on mac (win/linux should be similar) and blender 2.83.6 assuming you have blender in /Applications/Blender.app:**
get pip and install cython using pip
cd /Applications/Blender.app/Contents/Resources/2.83/python/bin/
./python3.7m -m ensurepip
./python3.7m -m pip install -U pip
./pip3 install cython
clone blender source
cd ~/Library/Application Support/Blender/2.83/scripts/addons/io_mesh_fast_obj
git clone git://git.blender.org/blender.git
cd blender
git checkout v2.83.6
build cython extension
/Applications/Blender.app/Contents/Resources/2.83/python/bin/python3.7m setup.py build_ext --inplace
***
## [Import Agisoft PhotoScan Cameras](https://github.com/uhlik/bpy/tree/master/io_import_photoscan_cameras.py)
Import cameras from Agisoft PhotoScan xml. Works with xml version 1.4.0 which is exported from PhotoScan 1.4.x versions and xml versions 1.5.0 from Agisoft Metashape 1.5.x versions. If you want to have images actually aligned with model, undistort images first. This is done in PhotoScan by `Export > Undistort Photos..`. Because you can't in Blender set resolution for cameras independently, xml with different cameras or image resolutions might not work well.
usage:
1. go to `Properties > Scene > Import Agisoft PhotoScan Cameras` panel
2. **Cameras XML**: set path to xml
3. set import options:
* **Camera Display Size**: size of imported cameras in viewport
* **Load Camera Images**: load images or not
* **Images Directory**: path to directory with undistorted images
* **Image Extension**: images extension, they all should be the same (currently)
* **Alpha**: camera image alpha, 0.0 - 1.0
* **Depth**: camera display depth, front / back
4. there are some more optional properties:
* **Create Chunk Region Borders**
* **Align to Active Object**: if you import mesh from PhotoScan first, the transform it to correct size and orientation, this option will copy transformation from that mesh if it is active
5. hit **Import**
6. import done..
7. now you can quickly swap cameras in alphabetical order in `PhotoScan Cameras Utilities` panel
changelog:
* 0.1.3 ui
* 0.1.2 compatibility with Agisoft Metashape XML (1.5.x)
* 0.1.1 first release
[BlenderArtist.org thread](https://blenderartists.org/t/addon-import-agisoft-photoscan-cameras/1140610)
***
## [Carbon Tools](https://github.com/uhlik/bpy/tree/master/carbon_tools.py)
Ever-evolving set of small tools, workflows and shortcuts focused mainly on processing photogrammetry scans.
#### Subtools
* **Extract** selected part of mesh to a new object. If edges of extracted mesh are changed, it won't be able to merge back seamlessly - use option to hide edges (lock button) to protect them.
* **Insert** it back when finished editing.
* **Extract Non-Manifold** elements with part of mesh around them (10x expanded selection) as subtool
#### Dyntopo
* **Dyntopo Setup** - Quick setup for optimizing mesh resolution. Set desired **Constant Resolution** and **Method** and hit **Dyntopo Setup**
Mode is switched to Sculpt with Dyntopo, brush is set to strength 0 - affecting only mesh resolution.
* **Dyntopo Live Settings** - current settings which can be changed during sculpting
#### Texture Paint
* **Texture Paint Setup** - Quick setup for retouching texture in Photoshop, set **Resolution** of exported images and hit **TP Setup** (Photoshop must be set in preferences as Image Editor)
* **External TP Live Commands**: **Quick Edit** - export image and open in PS, **Apply** - project image back to model, **Save All Images** - save all edited textures
#### IO
* **Import from ZBrush** (depends on other addon Fast Wavefront^2)
* **Export to ZBrush** (depends on other addon Fast Wavefront^2)
* **Transformation: Selected > Active** - Copy transformation from selected to active. Useful for setting correct scale and orientation after initial import from PhotoScan.
* **Matrix: Selected > Active** - Select matrix source object first and then target object. Matrix will be copied while keeping visual transformation intact.
* **Export to PhotoScan** (depends on other addon Fast Wavefront^2)
#### End
* **End** ends current procedure: Dyntopo and Texture Paint back to Object mode and reset all settings
#### Utilities
* **Smooth** / **Flat** shading, just shortcuts
* **UV Coverage** calculate and print how much percent covers active uv layout
* **Seams From Islands** mark seams from UV islands
* **Select Seams** select seam edges
* **Seams > Wireframe** copy seams edges to a new mesh object
* **Export UV Layout**
* **Wireframe** set display to shaded + wire + all edges and deselect object
* **Select Non-Manifold** select non-manifold elements and optionally focus camera on them (eye icon)
#### Conversions
* **UVTex > VCols** - Copy image colors from active image texture node in active material using active UV layout to new vertex colors
* **Group > VCols** - Active vertex group to new vertex colors, vertex weight to rgba(weight, weight, weight, 1.0)
* **VCols > Group** - Active vertex colors to new vertex group, vertex weight by color perceived luminance
* **Difference > Group** - Calculate difference between two selected meshes and write as vertex group to active mesh. Selected is considered to be original, active to be modified. Objects should have the same transformation.
changelog:
* 2.2.2 ui changes, mesh difference to group, calculate uv coverage
* 0.2.1 uvtex / vcols / group conversions
* 0.2.0 first release
***
## [Time Tracker](https://raw.githubusercontent.com/uhlik/bpy/master/system_time_tracker.py) (for blender 2.80)
Simple time tracker inside blender. After you install and enable it, it will log loaded and saved files and time spent of file until it is saved. All ui is inside addon preferences.
Here you can enable / disable logging, clear data collected so far, set custom data path (.csv) and see short summary of collected data and open individual project directories in file browser. The project name is determined by directory name where the blend is. For example if you have project in directory named "MyNewOutstandingProject" and all blends are inside subdirectory "models", set level number to 1 and you will see project name in results. 0 is directory directly above blend, 1 is one directory above blend, and so on. If you are like me and all your projects have the same subdirectory structure, sent directory level and you are good to go.
changelog:
* 0.2.0 updated for 2.80
* 0.1.0 added simple ui
* 0.0.8 ui tweaks, more advanced options, minor bugfixes
* 0.0.7 fixed performance and sorting, added tracking of files which were closed without saving once a minute (can be enabled in preferences: check Track Scene Update)
* 0.0.6 first release
[BlenderArtist.org thread](http://blenderartists.org/forum/showthread.php?345129-Time-Tracker-addon)