Of all the tutorials i have seen on nomad this was the most concise, and i appreciate it! You took a 3hour college lecture/demo and turned it into 3 minutes, well done 👏
You can also set the translation in the Gizmo Matrix like a slider - press and hold on the value and drag left / right...so you can already adjust the hightmap a bit in Nomad Sculpt. Nice tutorial.
Thank you Andrew. And no I did not know about the anaglyph generation capabilities of Blender. Thank you. I am fairly knew to 3D lenticular images. I’ve only done one really and that too, I handed off my depth map and original image it pertained to to someone else who generated the necessary frames, interlaced everything and printed. Between my depth map and print I have been using a Demo version of a software called 3D Masterkit. It’s somewhat popular in the world of Lenticular. Their support is quite responsive and when I don’t know how to do something, I’ll ask. I’m glad you know nodes! There are so many, that it is a bit intimidating on what to use and where to place them. Even though Nomad Sculpt is much faster at generating a depth map, perhaps using these nodes as you illustrated will allow me greater control? Alternatively, I could take the depth map made through Nomad Sculpt, and adjust it in photoshop…which I’ve done…but somehow the transitions to the different greyscale areas of my depth map don’t look as smooth as what I see when using the nodes in blender. I’ve been asking Chatgpt about the concept of masking in some way the area of my depth map that I want to have the node settings pertain to, but perhaps the RGB node suggestion you just gave will do the trick (because I think it is primarily the very white foreground…of the subjects shins that need to be toned down). Below is a Blender generated depth map of my image. I don’t think I adjusted this much, but it allows you to see the problem I am facing. I’ll send an anaglyph of how this depth map translates too. www.dropbox.com/s/jyo7vv6u48768vd/Prelim%20Depth%20for%20RP.jpg?dl=0 www.dropbox.com/s/ijz8b3l5vuqdg7o/Photo%20May%2002%202023%2C%209%2016%2019%20AM.jpg?dl=0
Thank you for sharing the images of your project. I do see what you are talking about. Using the RGB node to compress the range of the output image (as in my last reply) should get you what you are describing for the depth map. You can do something similar in Photoshop by adjusting the levels or curves (with either the Blender or Nomad Sculpt output). Looking at the 3DMasterKit documentation, you might also want to experiment with the “depth map factor” (described here: triaxes.com/docs/3dmk10/3DMasterKit-en/27Depthmapofthelayer.html ). I haven’t used 3DMasterKit at all but, my expectation is that setting the depth map factor below 1 would help with the issue you are having.
Hello Andrew, Thank you for this video. I have a sculpt that has about 1 million polygons. I have not retpologized it. I want to create a high resolution depth map. Similar to the one I see when I started watching your video above. I wanted to ask you if you feel I would need to retopologize my sculpt first in order to achieve similar quality results in Nomad Sculpt..or if I can just create a depth map from my raw sculpt? I do not yet have nomad sculpt, but I am considering it. Thank you for any insights you can share.
Greetings! These knots are around 150-200k polygons but, Nomad Sculpt can handle objects or scenes with at least 10 million polygons. Performance depends in part on your hardware and, I haven’t seen your model but, I would expect it to be okay without retopologizing. Note that you can also create depth maps in Blender. I cover that in my video previous to this one. Here is a direct link to the timecode: ruclips.net/video/vwuBbdCD0L0/видео.html
@@EvermoreStudio thank you Andrew. I will look at your video on how to make the depth map in Blender this weekend. And thank you for letting me know about your thoughts about nomad sculpt and it’s ability to handle my sculpt and it’s polygon count. In your link to the video, would I be able to make a depth map from a raw sculpt in Blender? Meaning, I wouldn’t need to retopologiz and I wouldn’t need to first render it before starting the depth map process in Blender (sorry if that is a dumb question, but I’m new)? Also, can Nomad Sculpt be used in any fashion with a PC? I don’t have a iPad (just and iPhone 11 Pro) and I am worried that the smaller screen (5.8”) might be a frustration. Perhaps if I can sculpt and see my progress on a PC it would be easier. Thank you for sharing your knowledge. I already feel empowered!
@@SaruPrasad You shouldn’t have to retopologize or do a full render to generate a depth map in blender. You may have to adjust some parameters to get a good depth map with your specific model. I think these are mostly shown in the video segment. Nomad Sculpt currently runs on iOS and Android. I know some people are sculpting with it on their phones. I am using a 10.9” iPad. The developer is considering making a Windows version but, it is not clear when that might happen. Nomad Sculpt seems to run ok even on older generation hardware. Depending on your financial situation, you might be able to find an inexpensive used iPad somewhere. Importing an existing model (e.g., created in Blender or, your preferred PC sculpting app) and, generating a depth map might be less difficult than trying to sculpt on a small screen. I’m glad I can help! I hope you are able to get what you need. I would love to see what you make, if you end up sharing it anywhere.
Hello again! I am not sure why your replies aren’t showing up in the comments but, I did receive them. You said you were able to get depth maps both with Blender and with the trial version of Nomad Sculpt. You were asking about controlling/compressing the depth of field. If you already have this set up like my other tutorial for Blender, you can adjust the Gamma node to change the amount of contrast between the closest and farthest points in the depth map. Lowering the gamma should adjust as you are describing, to have less distance between the parts nearest to and farthest from the viewer. You can likely also adjust this on your output image from either Blender or Nomad Sculpt in an image manipulation program (GIMP, Photoshop, etc.) to compress the dynamic range. Sometimes, there will also be an option in whatever software you are using to produce your final output that allows you to adjust that. The Stereogram generator I used in that video has a “Depth” parameter that lets you adjust how it translates in the output.
Yes. There is more detail on doing exactly that in the previous video. I have only done it with these knotwork models so far but, it should be similar for anything else.
Hello Andrew, hope that you are doing well. I followed your Blender tutorial to make a depth map. It came out well. However it is too contrasty. I adjusted the gamma node, but I still feel it is too contrasty. Would it be advisable/possible to add a second gamma node to the tree to gain additional gamma adjustment capabilities? Do you have cyan/red glasses for viewing anaglyphs. If so, I could show you how the contrast is translating to my 3D Image I’m creating. If there was some way of adjusting (perhaps the nodes…which I know nothing about beyond following your video) on the front end to more closely depict the true depth…that would save me time on the back end from needing to tweak it. The ultimate would be to have the depth map mirror the depth of my actual 3D sculpt. And for it (the depth map) to try to represent that depth within a narrow range of brightness levels. Right now I believe the range is something like RGB 200,200,200 to RGB 40,40,40. Thank you again Andrew for any ideas you can share. You’ve been quite helpful.
Hello again! I have done a fair amount of stereo photography. So, I do have red-blue glasses and other viewers around if you want to share an image. Presumably, you already know about Blender’s built-in stereoscopic output (anaglyph, stereo pairs, etc.) and, those aren’t suitable for your lenticular prints ( docs.blender.org/manual/en/2.80/render/output/multiview/usage.html ). For the lenticular prints I played with, I started with stereo pairs and, interleaved vertical strips from them that matched the lens sheet. That was quite some time ago. I imagine the tools are completely different now. If you can’t get the flatness you want out of the gamma, you might try some of the other adjustment nodes in the “color” submenu. You could, for example, use an RGB Curves node between the Gamma and the Viewer nodes (connecting the inputs and outputs named Image) or, replacing the Gamma node entirely. Then, adjust the curve to completely flatten off the highlights and/or dark values (e.g., move the point in the upper right corner straight down the right wall to make a very shallow ramp - the flatter it is, the less range the grayscale image will have). What other software are you using between the depth map and the final print? Everything I have worked with that takes a depth map for input has had some way to adjust how the values are translated into depth in the final output.
Kind of. You can make a custom alpha stamp from a depth map, then apply it to a high enough resolution mesh object in Nomad Sculpt. Of course, you will get a model cut off flat on the back side. There some folks with more expertise who might have might have better suggestions, like @ProcreateFX here or on the Nomad Sculpt forum. I think there are some people playing with using stuff like Stable Diffusion in a way that might yield models from depth maps but, I haven’t done anything with that.
Of all the tutorials i have seen on nomad this was the most concise, and i appreciate it! You took a 3hour college lecture/demo and turned it into 3 minutes, well done 👏
Love it! Great job. Just what I needed!
You can also set the translation in the Gizmo Matrix like a slider - press and hold on the value and drag left / right...so you can already adjust the hightmap a bit in Nomad Sculpt.
Nice tutorial.
Thank you! I should have mentioned that.
Thank you 👍
Thank you Andrew. And no I did not know about the anaglyph generation capabilities of Blender. Thank you.
I am fairly knew to 3D lenticular images. I’ve only done one really and that too, I handed off my depth map and original image it pertained to to someone else who generated the necessary frames, interlaced everything and printed.
Between my depth map and print I have been using a Demo version of a software called 3D Masterkit. It’s somewhat popular in the world of Lenticular. Their support is quite responsive and when I don’t know how to do something, I’ll ask.
I’m glad you know nodes! There are so many, that it is a bit intimidating on what to use and where to place them.
Even though Nomad Sculpt is much faster at generating a depth map, perhaps using these nodes as you illustrated will allow me greater control? Alternatively, I could take the depth map made through Nomad Sculpt, and adjust it in photoshop…which I’ve done…but somehow the transitions to the different greyscale areas of my depth map don’t look as smooth as what I see when using the nodes in blender.
I’ve been asking Chatgpt about the concept of masking in some way the area of my depth map that I want to have the node settings pertain to, but perhaps the RGB node suggestion you just gave will do the trick (because I think it is primarily the very white foreground…of the subjects shins that need to be toned down).
Below is a Blender generated depth map of my image. I don’t think I adjusted this much, but it allows you to see the problem I am facing. I’ll send an anaglyph of how this depth map translates too.
www.dropbox.com/s/jyo7vv6u48768vd/Prelim%20Depth%20for%20RP.jpg?dl=0
www.dropbox.com/s/ijz8b3l5vuqdg7o/Photo%20May%2002%202023%2C%209%2016%2019%20AM.jpg?dl=0
Thank you for sharing the images of your project. I do see what you are talking about.
Using the RGB node to compress the range of the output image (as in my last reply) should get you what you are describing for the depth map. You can do something similar in Photoshop by adjusting the levels or curves (with either the Blender or Nomad Sculpt output).
Looking at the 3DMasterKit documentation, you might also want to experiment with the “depth map factor” (described here: triaxes.com/docs/3dmk10/3DMasterKit-en/27Depthmapofthelayer.html ). I haven’t used 3DMasterKit at all but, my expectation is that setting the depth map factor below 1 would help with the issue you are having.
Hello Andrew,
Thank you for this video. I have a sculpt that has about 1 million polygons. I have not retpologized it. I want to create a high resolution depth map. Similar to the one I see when I started watching your video above. I wanted to ask you if you feel I would need to retopologize my sculpt first in order to achieve similar quality results in Nomad Sculpt..or if I can just create a depth map from my raw sculpt?
I do not yet have nomad sculpt, but I am considering it. Thank you for any insights you can share.
Greetings! These knots are around 150-200k polygons but, Nomad Sculpt can handle objects or scenes with at least 10 million polygons. Performance depends in part on your hardware and, I haven’t seen your model but, I would expect it to be okay without retopologizing.
Note that you can also create depth maps in Blender. I cover that in my video previous to this one. Here is a direct link to the timecode:
ruclips.net/video/vwuBbdCD0L0/видео.html
@@EvermoreStudio thank you Andrew. I will look at your video on how to make the depth map in Blender this weekend. And thank you for letting me know about your thoughts about nomad sculpt and it’s ability to handle my sculpt and it’s polygon count.
In your link to the video, would I be able to make a depth map from a raw sculpt in Blender? Meaning, I wouldn’t need to retopologiz and I wouldn’t need to first render it before starting the depth map process in Blender (sorry if that is a dumb question, but I’m new)?
Also, can Nomad Sculpt be used in any fashion with a PC? I don’t have a iPad (just and iPhone 11 Pro) and I am worried that the smaller screen (5.8”) might be a frustration. Perhaps if I can sculpt and see my progress on a PC it would be easier. Thank you for sharing your knowledge. I already feel empowered!
@@SaruPrasad You shouldn’t have to retopologize or do a full render to generate a depth map in blender. You may have to adjust some parameters to get a good depth map with your specific model. I think these are mostly shown in the video segment.
Nomad Sculpt currently runs on iOS and Android. I know some people are sculpting with it on their phones. I am using a 10.9” iPad. The developer is considering making a Windows version but, it is not clear when that might happen. Nomad Sculpt seems to run ok even on older generation hardware. Depending on your financial situation, you might be able to find an inexpensive used iPad somewhere.
Importing an existing model (e.g., created in Blender or, your preferred PC sculpting app) and, generating a depth map might be less difficult than trying to sculpt on a small screen.
I’m glad I can help! I hope you are able to get what you need. I would love to see what you make, if you end up sharing it anywhere.
Hello again! I am not sure why your replies aren’t showing up in the comments but, I did receive them. You said you were able to get depth maps both with Blender and with the trial version of Nomad Sculpt. You were asking about controlling/compressing the depth of field. If you already have this set up like my other tutorial for Blender, you can adjust the Gamma node to change the amount of contrast between the closest and farthest points in the depth map. Lowering the gamma should adjust as you are describing, to have less distance between the parts nearest to and farthest from the viewer. You can likely also adjust this on your output image from either Blender or Nomad Sculpt in an image manipulation program (GIMP, Photoshop, etc.) to compress the dynamic range.
Sometimes, there will also be an option in whatever software you are using to produce your final output that allows you to adjust that. The Stereogram generator I used in that video has a “Depth” parameter that lets you adjust how it translates in the output.
Will this make it possible to create bas reliefs? For instance, use a perspective view of 3D models and export depth map?
Yes. There is more detail on doing exactly that in the previous video. I have only done it with these knotwork models so far but, it should be similar for anything else.
Hello Andrew, hope that you are doing well. I followed your Blender tutorial to make a depth map. It came out well. However it is too contrasty. I adjusted the gamma node, but I still feel it is too contrasty.
Would it be advisable/possible to add a second gamma node to the tree to gain additional gamma adjustment capabilities? Do you have cyan/red glasses for viewing anaglyphs. If so, I could show you how the contrast is translating to my 3D Image I’m creating.
If there was some way of adjusting (perhaps the nodes…which I know nothing about beyond following your video) on the front end to more closely depict the true depth…that would save me time on the back end from needing to tweak it.
The ultimate would be to have the depth map mirror the depth of my actual 3D sculpt. And for it (the depth map) to try to represent that depth within a narrow range of brightness levels. Right now I believe the range is something like RGB 200,200,200 to RGB 40,40,40.
Thank you again Andrew for any ideas you can share. You’ve been quite helpful.
Hello again!
I have done a fair amount of stereo photography. So, I do have red-blue glasses and other viewers around if you want to share an image.
Presumably, you already know about Blender’s built-in stereoscopic output (anaglyph, stereo pairs, etc.) and, those aren’t suitable for your lenticular prints ( docs.blender.org/manual/en/2.80/render/output/multiview/usage.html ). For the lenticular prints I played with, I started with stereo pairs and, interleaved vertical strips from them that matched the lens sheet. That was quite some time ago. I imagine the tools are completely different now.
If you can’t get the flatness you want out of the gamma, you might try some of the other adjustment nodes in the “color” submenu. You could, for example, use an RGB Curves node between the Gamma and the Viewer nodes (connecting the inputs and outputs named Image) or, replacing the Gamma node entirely. Then, adjust the curve to completely flatten off the highlights and/or dark values (e.g., move the point in the upper right corner straight down the right wall to make a very shallow ramp - the flatter it is, the less range the grayscale image will have).
What other software are you using between the depth map and the final print? Everything I have worked with that takes a depth map for input has had some way to adjust how the values are translated into depth in the final output.
can you do this in reverse? bring a depth map in and create a model from the depth map?
Kind of. You can make a custom alpha stamp from a depth map, then apply it to a high enough resolution mesh object in Nomad Sculpt. Of course, you will get a model cut off flat on the back side.
There some folks with more expertise who might have might have better suggestions, like @ProcreateFX here or on the Nomad Sculpt forum.
I think there are some people playing with using stuff like Stable Diffusion in a way that might yield models from depth maps but, I haven’t done anything with that.
👍🏻