PDA

View Full Version : Creating Image Sets



LordEntrails
May 9th, 2024, 16:28
Fantasy Grounds version 4.5 introduced three image types for all Characters and NPCs. These images are:


a “Portrait” for Characters and called a “‘Full Body” for an NPCs (commonly called ‘portrait’)
a “Token-Flat” (aka token),
a “Token-Camera” (aka ‘full frontal standup’, aka ‘standup’)

This outlines how you can use a collection of applications to create sets of these images so that the portrait, token and standup are all of the same 3D model.
Applications Needed


HeroForge website account www.heroforge.com (http://www.heroforge.com/) (or another 3D modeler of your choice)
Gimp installed (or other image editing program) GIMP - Downloads (https://www.gimp.org/downloads/)
TokenTool (or other token creating or image editing tool) Download – RPTools (https://www.rptools.net/toolbox/download-rptools-products/)
FG Image Metadata Builder from Pre-defining the scale and positioning of images in 3D space (fantasygrounds.com) (https://www.fantasygrounds.com/forums/showthread.php?80375-Pre-defining-the-scale-and-positioning-of-images-in-3D-space)

Create Your Model

Create an account on www.heroforge.com (http://www.heroforge.com/)
Under Profile > Options disable 3D Environment

Create your miniature (steps not detailed here)
Make sure to remove any base (Stage > Base)
Rename them to something intelligent, Save

Take 3 screen shots, insure mini is as large as possible without overlapping other parts of the user interfaceZoom and position for a portrait, take a roughly square screen shot

Zoom and position for top down, take screen shot

Zoom and position for a full frontal standup view, pay specific attention to the feet, take screen shot

Rename your screenshots intelligently.
For example:


Portrait “Kobold Dagger”
Top Down “Kobold Dagger token”
Full frontal “Kobold Dagger standup”


Edit Your Screen Shots

Note, I use Gimp and TokenTool so those are the command names I will use, but other image editing programs can be used as well.
Be aware of the image guidelines; Developer Guide - Product Guidelines - Fantasy Grounds Customer Portal - Confluence (atlassian.net) (https://fantasygroundsunity.atlassian.net/wiki/spaces/FGCP/pages/2037547009/Developer+Guide+-+Product+Guidelines#Development---Product-Image-Guidelines)
Edit the Portrait Image

Load image in Gimp
Edit Image > Canvas Size to be square. Move canvas to crop desired portion of image.

Image > Rescale Image to 256 x 256 pixels. Note, the FG guidelines indicate 100x100, but I keep the images larger because file sizes are still acceptable and to allow for future resolution updates.

Export As .webp file

Note, I accept the default 90 quality. For these small images the savings by setting to 75% hasn’t seemed necessary.
Edit the Full Frontal Standup Image

Load image in Gimp
Remove background by using ‘Fuzzy Select’ tool (u) to select and then delete background parts.

Crop to Content both to remove unneeded file size, but to insure background elements were removed.

Verify image is less than 1000 pixels in width and height Image > Image Porperties. Rescale if needed.
Export As .webp file
Edit the Top Down Token Image

Load image in Gimp

Remove background by using ‘Fuzzy Select’ tool (u) to select and then delete background parts.
Crop to Content both to remove unneeded file size, but to insure background elements were removed.
Export As .webp file
Load image in Token Tool

Select desired Overlay
For small and medium tokens, set size to 256 x 256
Recommended Settings: Send to Back – Enabled, Clip Portrait – Disabled

Scale, Rotate & position as desired. Use the top right preview window to see what the export will look like.

File > Save As and make sure location, name and extension are correct (i.e. “Kobold Dagger token.webp”
Adding XML Data to Standup Image

Open the FG Image Metadata Builder application
Load your standup image
Create XML file per: Developer Video FG Image Metadata Tool (youtube.com) (https://www.youtube.com/watch?v=7Y_BqXN1r6Q)
Packaging Your Images

Because FG handles portraits, tokens, and standups differently, you will end up creating both a Portrait Pack (.ppk) and Asset/Toekn Module (.mod) files. This results that in FG:


Portraits can be accessed by the portrait selection tools (Portrait filter in the Assets tool, and through double clicking the portrait image).
Tokens can be accessed by the Tokens filter in Assets.
Standup can be accessed by the Images filter in Assets (and if you name as suggested here, via ‘standup’ search).

Create a Thumbnail Image

Using the images, create a 300x300 PNG image to use as a thumbnail. Such as:
Create the Token and Standup Module

The FG wiki article is out of date (as of 1 May 2024) in one respect, it does not have information on what to do with the standup/token-camera images. Follow the Module - Token Pack Creation - Fantasy Grounds Customer Portal - Confluence (atlassian.net) (https://fantasygroundsunity.atlassian.net/wiki/spaces/FGCP/pages/996644379/Module+-+Token+Pack+Creation) wiki article except as noted below;


Create an additional folder called “images”. Place all the standup images and XML files in there and then include that folder when zipping the folders to make the .mod file.
Remove the ‘token’ from the token file name once the file is in the tokens folder. (I leave 'standup' so they can be searched for in mass in FG.)

Note: Make sure the folders only contain the image files and not any “desktop.ini” or other operating system files.
Create the Portrait Pack

Follow the directions here: Module - Portrait Pack Creation - Fantasy Grounds Customer Portal - Confluence (atlassian.net) (https://fantasygroundsunity.atlassian.net/wiki/spaces/FGCP/pages/996644390/Module+-+Portrait+Pack+Creation)
Note: Make sure the folders only contain the image files and not any “desktop.ini” or other operating system files.

See attachments for instructions with screenshots for most steps.

Laerun
May 9th, 2024, 17:39
Thanks for this, very useful.

wndrngdru
May 10th, 2024, 23:50
If you're okay with its terms of use (I didn't notice anything inherently nefarious), there is a useful website that uses AI to remove backgrounds. and it works surprisingly well.

https://www.remove.bg/

On occasion, I'll need to pull the output into GIMP afterward to clean something it missed, but that usually only happens where the character and background are nearly indistinguishable.