Creating textures from Ornate Photos - Revision history

Nbohr1more: /* AI Tools */

2024-12-17T01:34:14Z

AI Tools

← Older revision		Revision as of 01:34, 17 December 2024
Line 156:		Line 156:
	==== '''AI Tools''' ====		==== '''AI Tools''' ====

	<br>Shortly after creating this wiki, Arcturus ( team member ) demonstrated that there are several [https://huggingface.co/spaces/Stable-X/~~normal-estimation-arena~~ Stable Diffusion AI clients]		<br>Shortly after creating this wiki, Arcturus ( team member ) demonstrated that there are several [https://huggingface.co/spaces/Stable-X/StableNormal_turbo_beta Stable Diffusion AI clients]
	<br>that do a great job of estimating depth or normalmaps from photo sources.		<br>that do a great job of estimating depth or normalmaps from photo sources.
	<br>		<br>

2024-07-18T19:17:14Z

Tips and Tricks

← Older revision		Revision as of 19:17, 18 July 2024
Line 148:		Line 148:
	[[File:Awesome_girard_preview.png\|800px]]		[[File:Awesome_girard_preview.png\|800px]]

	~~Updated~~: Since the green color channel is where top-shading originates in AwesomeBump, it's probable best to only blend photo shading there.		Update: Since the green color channel is where top-shading originates in AwesomeBump, it's probable best to only blend photo shading there.
			<br>Using the top shading from the photo source in place of the green channel will only work for very direct illumination, sunlight or other larger
			<br>diffuse sources will have natural ambient occlusion on top of protrusions which will foil the normal shader. Unless you want to manually paint the
			<br>AO artifacts away or the scene is simple enough to use erode and blur filters, you'll probably want to mix generated normals into the channel.
	<br>Blue represents steepness, and can be approximated by using shading from both red and green brightness in both standard and inverted modes overlapped		<br>Blue represents steepness, and can be approximated by using shading from both red and green brightness in both standard and inverted modes overlapped

2024-07-17T16:13:10Z

AI Tools

← Older revision		Revision as of 16:13, 17 July 2024
Line 168:		Line 168:
	<br>so the AI only needs to analyze the detail of the cropped portion. Then just scale the results back down and overlay		<br>so the AI only needs to analyze the detail of the cropped portion. Then just scale the results back down and overlay
	<br>it on the macro-scale AI generated normalmap.		<br>it on the macro-scale AI generated normalmap.
	<br><br>Just as with njob or AwesomeBump it's best to [[#De-Noise_the_Image\|desaturate and de-noise the image]] prior to AI processing		<br><br>Just as with njob or AwesomeBump it's best to [[#De-Noise_the_Image\|desaturate and de-noise the image]] prior to AI processing ( but not necessary )
	<br><br>In many cases, this workflow can probably completely supplant the other methods outlined in this wiki and other similar references.		<br><br>In many cases, this workflow can probably completely supplant the other methods outlined in this wiki and other similar references.
	<br>		<br>

2024-07-17T14:51:37Z

Tips and Tricks

← Older revision		Revision as of 14:51, 17 July 2024
Line 149:		Line 149:

	Updated: Since the green color channel is where top-shading originates in AwesomeBump, it's probable best to only blend photo shading there.		Updated: Since the green color channel is where top-shading originates in AwesomeBump, it's probable best to only blend photo shading there.
	Blue represents steepness, and can be approximated by using shading from both red and green brightness in both standard and inverted modes overlapped		<br>Blue represents steepness, and can be approximated by using shading from both red and green brightness in both standard and inverted modes overlapped

	==== '''AI Tools''' ====		==== '''AI Tools''' ====

2024-07-17T14:51:10Z

Tips and Tricks

← Older revision		Revision as of 14:51, 17 July 2024
Line 147:		Line 147:

	[[File:Awesome_girard_preview.png\|800px]]		[[File:Awesome_girard_preview.png\|800px]]

			Updated: Since the green color channel is where top-shading originates in AwesomeBump, it's probable best to only blend photo shading there.
			Blue represents steepness, and can be approximated by using shading from both red and green brightness in both standard and inverted modes overlapped

	==== '''AI Tools''' ====		==== '''AI Tools''' ====

2024-07-17T13:40:01Z

Tips and Tricks

← Older revision		Revision as of 13:40, 17 July 2024
Line 187:		Line 187:

	<br>		<br>
	Results:		Results ( of everything basically: Awesomebump, Top shading, Mean Curvature Blur, Macro AI, Crop AI, and color channel inverts )

	[[File:Girard_local_aimerge5.png\|600px]]		[[File:Girard_local_aimerge5.png\|600px]]

2024-07-17T12:42:49Z

Invert the Green Channel

← Older revision		Revision as of 12:42, 17 July 2024
Line 229:		Line 229:
	[[File:Girard_invert.png\|600px]]		[[File:Girard_invert.png\|600px]]
	<br><br>See also: [[Inverting_Normalmaps\|Inverse Normalmaps]]		<br><br>See also: [[Inverting_Normalmaps\|Inverse Normalmaps]]
			{{clear}}
			<br>

	== '''Creating the Diffuse''' ==		== '''Creating the Diffuse''' ==

2024-07-17T12:41:53Z

Invert the Green Channel

← Older revision		Revision as of 12:41, 17 July 2024
Line 228:		Line 228:
	<br>		<br>
	[[File:Girard_invert.png\|600px]]		[[File:Girard_invert.png\|600px]]
	<br>See also: [[Inverting_Normalmaps\|Inverse Normalmaps]]		<br><br>See also: [[Inverting_Normalmaps\|Inverse Normalmaps]]

	== '''Creating the Diffuse''' ==		== '''Creating the Diffuse''' ==

2024-07-17T12:41:32Z

Invert the Green Channel

← Older revision		Revision as of 12:41, 17 July 2024
Line 228:		Line 228:
	<br>		<br>
	[[File:Girard_invert.png\|600px]]		[[File:Girard_invert.png\|600px]]
			<br>See also: [[Inverting_Normalmaps\|Inverse Normalmaps]]

	== '''Creating the Diffuse''' ==		== '''Creating the Diffuse''' ==

2024-07-17T05:00:11Z

AI Tools

← Older revision		Revision as of 05:00, 17 July 2024
Line 185:		Line 185:
	# Merge the layers and export		# Merge the layers and export
	# Preview the results in AwesomeBump, if it looks good invert the green channel and test in TDM		# Preview the results in AwesomeBump, if it looks good invert the green channel and test in TDM

			<br>
			Results:

			[[File:Girard_local_aimerge5.png\|600px]]

	<br>A quicker / easier way to make use of the AI generated "macro scale" image is to use the addnormal material keyword in your [[#Material_Def_and_Paths\|material def]]:		<br>A quicker / easier way to make use of the AI generated "macro scale" image is to use the addnormal material keyword in your [[#Material_Def_and_Paths\|material def]]: