This is the first tutorial I come across on this channel which manages to explain everything in a clear way from A to Z. Most other tutorials are far too complicated for a beginner. I hope Capturing Reality rethinks its pedagogical approach. It's really a great product but the UI and the contents are quite overwhelming.
Pls make CLI video also. This video was SOOOOO helpful. I love the software but most of the the TH-cam tutorials were of the older version and only helped about 80-85%
This is by VERY far the best photogrammetry software I have tried so far. I just wish it would play nice with my 3Dconnexion mouse... I just hate the clickety-clicking needed to move around with mouse + keyboard.
Oh, i was so excited when i see there was an unreal photogrammetry software "for free" and then I looked at this tutorial and then i saw the "pay to render the scan" i got disapointed, i hate when people or companies say "free" but then there is a payment after you download and start using it.
Welp now I feel much more comfortable in getting into photogrammetry. Just curious will it ever be possible to capture reflective or translucent surfaces? I guess it would require light field capturing or something?
Amazing tutorial. I was able to follow along pretty well even with the newer 1.4V. I did all the work and when I did the alignment the second time it said " unsupported camera model and the calibration parameters is not supported. why the heck didnt it tell me that an hr and a half ago and before the account set up. what a pain in the ass.
Would love to see a preprocessing masking feature similar to metashape. I have wasted some $ creating masks for complex objects that were never able to merge.
Amazing tutorial!!! When we get the final textured model, can we export into quixel Mixer to fine tune the texture even further for the purpose of a detailed asset for a game? (New to modeling/development sorry!)
I hope a Linux version of this is made! This is the only software that I need WIndows for now... UnReal Engine is coming out with a Linux version of that engine.. hopefully it can be done with RC as well! :)
As far as I understand, you can't reach the end result without saving the previous two scans? And I will be paying credits again for saving the final result?
Hi, 1.0 shouldn't work anymore. In 1.2 you can find it under Scene 3D tab/Tools/Mesh Model. Lassos for cameras and sparse point cloud are under Scene 3D tab/Tools/Registration & Sparse Point Cloud
Pavel, that's my setup..... More ram would be more confortable I think.
3 ปีที่แล้ว
@@florentthouret2564 I'm sitting on 64GB ram system now and it only gets fully used when I get to 500+ 24megapixel pics & 3x16K textures range... RC is using comparably less RAM than some other apps (which is not necessary good thing)
@ I'm running Intel(R) Core(TM) i7-4790S CPU @ 3.20GHz -3.20 GHz, 16GB DDR3 system RAM, and a GeForce GTX 4GB GDDR6 graphics card, I've noticed my results aren't the greatest
3 ปีที่แล้ว
@@digital-designferrante336 I'm assuming you are running an SSD. Dual nvme is great help too. Better cpu+gpu would definitely help, but we are talking about 10minutes differences in this ~150 pics range. RC 1.2 Tarrasque is very fast. So it all depends on what on your goals and how many models you need per day.
Unfortunately, we do not support 360 degree images. This is a popular feature request, so I will add your comments to our system of feature requests and this will up its priority.
Is the screen without any not movable features? Is the only thing visible on the images only the captured object? How many images were used (overlap, angles)? What are your alignment settings? Have you used masks?
when i try to texturize the model... it says to increase the triangle count... how do i do that... my model has alot of photos in it and the memory is not enough.
@@CapturingReality when i try to texturize my model, it says ... INCREASE MAXIMAL TEXTURE COUNT OR MAXIMAL TEXTURE RESOLUTION. how do i do this ?.the caps is exactly what it says.. maybe their are to many photos, is it important to texturize the model.
@@gladiatormechs5574 You can do this in the texture settings under Mesh Model tab/Mesh Color & Texture/Settings/Defaul unwrap parameters and there you have options for changing the Maximal texture resolution and Maximal texture count. You can try to change the maximal texture count to a higher number.
Hi François, coin could be tricky and it mainly depend how it is captured. The hardest part is the thin part of the coin. There is possibility to use control points to merge two separated part of the coin. How did you capture the images?
@@CapturingReality Thank you for your answer ! I use a turntable for the coin. It rotates 15° around its central axis. The camera is fixed on a ball joint in order to move from 90° to 30° by 15°. 100 pictures by face. I want to use Reality Capture for all processing...
@@phoebussolutions If you are sure that the camera positions are still the same, you can try to use XMP workflow. You will need to get the camera positions and then use these positions to alignment. th-cam.com/video/5VyYLaNxHz0/w-d-xo.html
This is the first tutorial I come across on this channel which manages to explain everything in a clear way from A to Z. Most other tutorials are far too complicated for a beginner. I hope Capturing Reality rethinks its pedagogical approach. It's really a great product but the UI and the contents are quite overwhelming.
Every other video I've seen on this workflow is so complicated, your one was super straight forward, thanks!
Pls make CLI video also. This video was SOOOOO helpful. I love the software but most of the the TH-cam tutorials were of the older version and only helped about 80-85%
Please do a CLI step by step version that would be cool to see from start to finish. Great content!
yes please! Would love to learn that cheers!
@@aneltoromanovic3362 what is cli?
Would love to see clI content!
Thanks. That was a first rate introduction into Reality Capture
This is by VERY far the best photogrammetry software I have tried so far.
I just wish it would play nice with my 3Dconnexion mouse... I just hate the clickety-clicking needed to move around with mouse + keyboard.
Hi Olivier, thank you for nice words. I also added your vote for this feature request
@@CapturingReality Great ! I just hope it won't take years of begging like with Zbrush ! ;)
Very good introduction and quite useful RC tutorial!
This is a really helpful and well explained video. Thank you very much. I would be very interested in the CLI process.
Please do a CLI version on this topic! Great video!
Hi everyone thank you for the idea for next video. What do you want to see about CLI?
Beginner stuff like what type of work CLI is best for, then example of how to setup one of those tasks. Thanks so much for these great tutorials!
CLI idea: how to do a simple map with drone photos through CLI (unless this is bad idea, then maybe a video of why its bad idea :-) )
@@KevinNMurphy Thank you for the ideas
Oh, i was so excited when i see there was an unreal photogrammetry software "for free" and then I looked at this tutorial and then i saw the "pay to render the scan" i got disapointed, i hate when people or companies say "free" but then there is a payment after you download and start using it.
Hi Darien, it is free. This tutorial is older for old PPI model, which is no more relevant.
Yes, please do a video showing script automation
Would love to see clI content too, I thought we can use cli only with a real license.
Since the new version it is possible to use CLI also with PPI license.
Great content! Please make the CLI scripting video for the community. Thanks a lot!
Pleaseeeeee do a CLI step by step guide, much appreciated!
Hi, you can also watch this: th-cam.com/video/KzWW-j6Zjqc/w-d-xo.html
Welp now I feel much more comfortable in getting into photogrammetry. Just curious will it ever be possible to capture reflective or translucent surfaces? I guess it would require light field capturing or something?
As far as I know you need tools like AESUB Blue spray
@@florentthouret2564 I’ll look into it. Thank you.
matter spraying and cross-polarized setups are ways to go... but neither will be optimal for stuff like car windows
@ Thank you I’ll look into them
Amazing tutorial. I was able to follow along pretty well even with the newer 1.4V.
I did all the work and when I did the alignment the second time it said " unsupported camera model and the calibration parameters is not supported. why the heck didnt it tell me that an hr and a half ago and before the account set up. what a pain in the ass.
Hi, can you check this: forums.unrealengine.com/t/unsupported-camera-model-faro-focus-premium/1229437?
Отличное видео!
Ждём следующее с автоматизацией процесса.
Great tutorial. Thank you ❤
Wow that's great!
I'm curious how to get the sole of the shoe perfectly without hanging or levitating in the air. (I can't use my telekinesis...)
Hi, you can use mask like in this tutorial: th-cam.com/video/xxsdbzBwWLE/w-d-xo.html
Would love to see a preprocessing masking feature similar to metashape. I have wasted some $ creating masks for complex objects that were never able to merge.
Thank you for the idea. What object did you scan? Did you use control points in that project? What was used overlap?
Amazing tutorial!!! When we get the final textured model, can we export into quixel Mixer to fine tune the texture even further for the purpose of a detailed asset for a game? (New to modeling/development sorry!)
Yes, this is possible.
Let’s see cli way please!
Great video, an CLI video would be nice also 😊
I hope a Linux version of this is made! This is the only software that I need WIndows for now... UnReal Engine is coming out with a Linux version of that engine.. hopefully it can be done with RC as well! :)
Not yet, still in our feature request database.
As far as I understand, you can't reach the end result without saving the previous two scans? And I will be paying credits again for saving the final result?
No, you are just licensing used images/source data to create model. Then you can process those as many times as you wish without paying again.
How did you photograph the bottom of the model sitting on the table? I missed that bit.
It was just turned on the side and tehn aligned separate to create masks. Later masks were used and aligned all together.
*WHERE IS THE LASSO IN 1.0 VISION ?*
Hi, 1.0 shouldn't work anymore. In 1.2 you can find it under Scene 3D tab/Tools/Mesh Model. Lassos for cameras and sparse point cloud are under Scene 3D tab/Tools/Registration & Sparse Point Cloud
Yesss!!! I like it toooo!!!
Hi, is it possible to import RCP or RCS file into the reality capture?
Hi, it is not possible.
What would you recommend for system RAM and GPU RAM?
for smaller models like this, 16GB ram + any nVidia 4GB card will be more than enough
Pavel, that's my setup..... More ram would be more confortable I think.
@@florentthouret2564 I'm sitting on 64GB ram system now and it only gets fully used when I get to 500+ 24megapixel pics & 3x16K textures range... RC is using comparably less RAM than some other apps (which is not necessary good thing)
@ I'm running Intel(R) Core(TM) i7-4790S CPU @ 3.20GHz -3.20 GHz, 16GB DDR3 system RAM, and a GeForce GTX 4GB GDDR6 graphics card, I've noticed my results aren't the greatest
@@digital-designferrante336 I'm assuming you are running an SSD. Dual nvme is great help too. Better cpu+gpu would definitely help, but we are talking about 10minutes differences in this ~150 pics range. RC 1.2 Tarrasque is very fast. So it all depends on what on your goals and how many models you need per day.
Does this software handle 360 photos? Want to do the doll house effect using photos.
Unfortunately, we do not support 360 degree images. This is a popular feature request, so I will add your comments to our system of feature requests and this will up its priority.
Is it possible to reproject the texture on an external model imported in Reality Capture and then export the model using PPI license?
Yes, it is possible.
Alignment points are also under the object how did you do that, There are no tips on video.
The object was captured from both sides and there was used the mask's workflow. The workflow starts at th-cam.com/video/tw6wNNEbH_M/w-d-xo.html
please add possibility to use 360 cameras.
Hi, I will add this as a vote to our feature request database.
@@CapturingReality yes I do alot of 360 and want to see if this does doll house effect for vitural tours.
the alignment dosen't work with turntable and green screen, please help
Is the screen without any not movable features? Is the only thing visible on the images only the captured object? How many images were used (overlap, angles)? What are your alignment settings? Have you used masks?
sooooo,, helpful this video is, thanks. we need CLI also, wanna get it. CLI CLI CLI CLI CLI………… super thanks
Hi Michael, you can find Complete Model Using CLI tutorial at: th-cam.com/video/KzWW-j6Zjqc/w-d-xo.html
when i try to texturize the model... it says to increase the triangle count... how do i do that... my model has alot of photos in it and the memory is not enough.
What is the whole error message? Is it not about Maximal texture count?
@@CapturingReality when i try to texturize my model, it says ... INCREASE MAXIMAL TEXTURE COUNT OR MAXIMAL TEXTURE RESOLUTION. how do i do this ?.the caps is exactly what it says.. maybe their are to many photos, is it important to texturize the model.
@@gladiatormechs5574 You can do this in the texture settings under Mesh Model tab/Mesh Color & Texture/Settings/Defaul unwrap parameters and there you have options for changing the Maximal texture resolution and Maximal texture count. You can try to change the maximal texture count to a higher number.
Hello, I try to make a coin using the Mask tool and I can't merge the 2 faces ! Could you help me - Please ? Regards, François
Hi François, coin could be tricky and it mainly depend how it is captured. The hardest part is the thin part of the coin. There is possibility to use control points to merge two separated part of the coin. How did you capture the images?
@@CapturingReality Thank you for your answer !
I use a turntable for the coin. It rotates 15° around its central axis. The camera is fixed on a ball joint in order to move from 90° to 30° by 15°. 100 pictures by face. I want to use Reality Capture for all processing...
@@phoebussolutions If you are sure that the camera positions are still the same, you can try to use XMP workflow. You will need to get the camera positions and then use these positions to alignment. th-cam.com/video/5VyYLaNxHz0/w-d-xo.html
Super manuál neviem, kto dal dislike musim aj ja už nieco urobiť ale po Slovensky
i can make free in maya
发现了新大陆
啥玩意你发现了新大陆？
@@elen5110 当然是这款久违的超级赞"RealityCapture"!
Automate my life please