NICE - I think this is a really great way to create fast context models as well - especially if you can drop textures on from the photo, which we'll talk about in the future :)
No, i would like to ask, it you want to remake a toys like gundam, one photo is not enought to build. So we need a sample cube in real world. For example as bring some 10mm^3 Cubes and place them near the toy. Each angle of camera can see one of them but don't move them with different shot. So how to import multi photo to set all camera is fix position by those cube? and the cube size in 3d is same as the photo
A million thanks! Awesome teaching skills as usual! The texturing tutorial idea is also super welcome! Please consider offering paid courses on the subject so that we could have access to files, etc. Thanks again!
Thank you, thank you, thank you. I've learned so much in this 10 minute video. If you haven't already the tutorials you suggested at the end about texturing utilizing the picture and interior images would be wonderful.
Hi CG, I tried everything in the video but twice I am getting a Rhombus shaped base instead of Rectangular...what could possibly be the reason...from one angle ( picture ) it looks ok but when I rotate the model it gives away to rhombus.
I am attempting to model the Ohio State Reformatory where they filmed The Shawshank Redemption. I have an image of it that was shot really wide so it has vanishing points on both sides. More like a fisheye perspective. I only need to model half of the building so I'm working with the left side of the image hoping for the best. I have tried photogrammetry but would really need a drone to get a complete series of photos to do that properly. Since they built the new prison right next to the old one, you can imagine flying a drone anywhere near there is not an option. I need to make this model because I tried this before and couldn't get a decent result. So I can't leave it alone until I figure it out. It's just a challenge for me at this point. I already hand traced a vector image to feed into my CNC. Cutting it into 4 foot stair treads, engraving it on old glass mirrors and acrylic sheets. Really appreciate all the time and effort you put into making these videos. TH-cam has become the largest classroom the world has ever seen. I know from experience that I can learn how to use software, fix an appliance, pour a slab etc.. I remember when Google bought YT. I wondered what the hell they wanted it for. Look at it now. Thanks again for helping to make this free service the most comprehensive kick ass classroom for people like me. People used to ask me "How do you know how to do that?" I used to say "They hid it in a book..." Now its just "TH-cam." The old reformatory is about 5 miles from my place.
I have been doing some interior modeling, but I have seen two problems with fspy: 1) the proportions are not right, for example the ceiling in real life is 2.40 Meters high, but when I use the measure tool in blender, it says that is 5 meters or more. 2) if I want to model a room using two pictures taken from different angles, in theory, I would not have to model again the scene, at least not all of it, because those pictures were taken in the same place, so it means that the position of the camera is the only thing that changed, everything else, like the elements modeled, should stay in the same place, but that is not happening. If any one has the answer for those two issues, it would be great to hear from you
Yeah, I have the same issues. This video may help, th-cam.com/video/704NgSWO5fk/w-d-xo.html A lot of the tutorials on this are not in-depth, they don't explain that only certain images work. In the video above this the creator found an image measured it, and inputted that info into fSpy. There is is a lack of proportion, that's when you model like using this fSpy, objects are stretched and scale incorrectly. I just try the fSpy addon since the past 2 days, it's it a little more than placing some lines and it coming out perfect. I'm going to begin studying up on cameras, Things like focal point, and scaling is important. When you take a picture with a camera the objects are no stretched, it s real life representation to it. Of course, you have to scale to make it correct. It is a very useful tool, it just has to be used correctly.
Might be a dumb question but would I be able to use the image and add items. So I’m interested in using an existing photo and simply adding rendered images. Would that be possible? Appreciate the advice.
Awesome video! id love to see more on this like adding textures, interiors, also if its possible to use multiple photos from different sides as reference.thanks!
Hi, thanks for the video. Anyhow, I miss one information, the SIZE/SCALE. My feeling is that somewhere we have to manually define the size, but I don't see. where we are doing it.
I figured it out meanwhile. There is a "Reference distance" option on the left side. There you select an appropriate axis, where you have some known size information abou the object. A new scale appears on that axis. You move its two ends to the measured objects edges. Also you write the measured distance in the dialog box. If you save the fSpy file and import it to Blender, it will use proper scale.
id like to know how to add the reference image onto the model i guessing uv wrap must come into play just want to get it as realistic to photo as possible
Nice. Was just wondering. After importing the fspy data into blender and adding whatever additional models into the scene (after doing all the camera mapping and stuff); would it then be possible to transfer all that info to unreal engine to continue work? Is there a way to transfer the fspy camera info to an unreal camera? Not sure if I’m explaining properly. I ask u cos I know u don’t restrict yourself to blender alone.
Does anybody else have an issue where they make the fspy camera view and import it, but the image won’t load in? The camera angle changes to the correct one, but I can’t import the image with it.
This is indeed a very helpful tool. I don't know much about the Blender universe, so I don't know of alternative ways that might produce similar results. *However*, I would find it much more productive if the necessary step 1, the generation of the perspective would not take place in a stand alone tool, but directly in the Blender addon. Can this not be done somehow?
Can you then map onto each wall the relevant piece of photo and then if you move the 3D barn to have the wall with doors face on will the photo texture change perspective accordingly?
Well a lot depends on the level of complexity - it's best for projects that have very defined perspective lines - curved/non-standard shapes this might not be the right tool
Is there anyone here that woukd be willing to make cgi images from some photos for me to use for a movie project I’m working on?? I have lower quality pictures that I’d like to be made into cgi images so I can use them for cgi backgrounds for green screen work. If anyone is interested in doing this, how much would you charge me to do this??
Funny how Blender Guru says when he started making videos he sounded terrible. "I sounded like this" then proceeds to sound exactly like Justin but just his starting part. That "Hey guys" part.
@@TheCGEssentials If you sounded the same when you explain in a video as you do when you start a video I would have nothing to say. When you start a video it sounds too intimate, that voice like you don't want to wake up anyone in the house and whisper in my ear kinda voice. Hey! Maybe it just me so just ignore this comment.
![[#2024MAMA] ROSÉ (로제), Bruno Mars - APT. | Mnet 241122 방송](http://i.ytimg.com/vi/dgGqD28J6aQ/mqdefault.jpg)
Hi everyone! What have you been modeling with this add-on? Let me know in the comments below! :)
NICE - I think this is a really great way to create fast context models as well - especially if you can drop textures on from the photo, which we'll talk about in the future :)
No, i would like to ask, it you want to remake a toys like gundam, one photo is not enought to build.
So we need a sample cube in real world. For example as bring some 10mm^3 Cubes and place them near the toy.
Each angle of camera can see one of them but don't move them with different shot.
So how to import multi photo to set all camera is fix position by those cube? and the cube size in 3d is same as the photo
I try some 3d scan app. But the quality is not enough for some like gundam toys detail.
You're a Boss. You never leave questions that arise unanswered. Your basic model; that your dog? For sure I think. Wonderful! Give her a big hug.
Blender is becoming more user friendly than Sketchup every time I see one of you videos. This is excellent! Thanks!
It is really simple I have been using it for just 2 weeks ans I have made some amazing things
Interesting - this would be the first time I've heard Blender described as more user friendly than SketchUp...
The CG Essentials Yeah....not quite there.......yet ;)
Thank you for the clear instruction on f-spy!. Great demo! very excited about this add-on
As others have pointed out, your teaching skills are top notch. Appreciate your contribution very much.
Thank you! Your tutorials are always really great. Your demeanor and teaching skills are top notch!
Glad you like them!
A million thanks! Awesome teaching skills as usual! The texturing tutorial idea is also super welcome! Please consider offering paid courses on the subject so that we could have access to files, etc. Thanks again!
Thank you, thank you, thank you. I've learned so much in this 10 minute video. If you haven't already the tutorials you suggested at the end about texturing utilizing the picture and interior images would be wonderful.
Thank you much, I was confused watching other tutorials, yours was straight to the purpose
Great tutorial.So many possibilties
As always, the most useful and fabulously explained. Thank you!
Dude thank you so much. I literally used a image with almost nothing I could place prespective lines on and it still worked really well
Glad it helped!
Exactly what I need, I will test it the next time. Thank you very much!
Impressive, I could imagine having a lot of fun recreating fictional locations from anime/cartoons as well.
YES - sounds fun to me :)
thanks i was looking for this addon and you helped me a lot
love your videos
Thanks for the tutotial...do you know how to adjust the brightness or transparency of the background image?
Its a placer watch and learn with you! Thank you for this tutorial!!
Glad you liked it!
Insanely helpful and simple tutorial, thank you so much.
Thanks so much, I kept having problems with installing the blender Add-on, until I saw your video.
Yeah this one's a bit tricky. Glad you got it! :)
This is what we want, Thanks
Hi CG, I tried everything in the video but twice I am getting a Rhombus shaped base instead of Rectangular...what could possibly be the reason...from one angle ( picture ) it looks ok but when I rotate the model it gives away to rhombus.
Brou this safe my life, thanks for the importer addon part that was a really good tip!
I am attempting to model the Ohio State Reformatory where they filmed The Shawshank Redemption. I have an image of it that was shot really wide so it has vanishing points on both sides. More like a fisheye perspective. I only need to model half of the building so I'm working with the left side of the image hoping for the best. I have tried photogrammetry but would really need a drone to get a complete series of photos to do that properly. Since they built the new prison right next to the old one, you can imagine flying a drone anywhere near there is not an option.
I need to make this model because I tried this before and couldn't get a decent result. So I can't leave it alone until I figure it out. It's just a challenge for me at this point. I already hand traced a vector image to feed into my CNC. Cutting it into 4 foot stair treads, engraving it on old glass mirrors and acrylic sheets.
Really appreciate all the time and effort you put into making these videos. TH-cam has become the largest classroom the world has ever seen. I know from experience that I can learn how to use software, fix an appliance, pour a slab etc.. I remember when Google bought YT. I wondered what the hell they wanted it for. Look at it now.
Thanks again for helping to make this free service the most comprehensive kick ass classroom for people like me.
People used to ask me "How do you know how to do that?" I used to say "They hid it in a book..."
Now its just "TH-cam."
The old reformatory is about 5 miles from my place.
Thank you men! That has totaly helpfull.
i had no idea fpsy could be used in that way, feels like you inverted the concept, when compared to what I had seen before. neat and interesting.
I assume what you'd seem before was more of an interior render?
I have been doing some interior modeling, but I have seen two problems with fspy:
1) the proportions are not right, for example the ceiling in real life is 2.40 Meters high, but when I use the measure tool in blender, it says that is 5 meters or more.
2) if I want to model a room using two pictures taken from different angles, in theory, I would not have to model again the scene, at least not all of it, because those pictures were taken in the same place, so it means that the position of the camera is the only thing that changed, everything else, like the elements modeled, should stay in the same place, but that is not happening.
If any one has the answer for those two issues, it would be great to hear from you
Yeah, I have the same issues. This video may help, th-cam.com/video/704NgSWO5fk/w-d-xo.html
A lot of the tutorials on this are not in-depth, they don't explain that only certain images work. In the video above this the creator found an image measured it, and inputted that info into fSpy. There is is a lack of proportion, that's when you model like using this fSpy, objects are stretched and scale incorrectly.
I just try the fSpy addon since the past 2 days, it's it a little more than placing some lines and it coming out perfect. I'm going to begin studying up on cameras, Things like focal point, and scaling is important. When you take a picture with a camera the objects are no stretched, it s real life representation to it. Of course, you have to scale to make it correct. It is a very useful tool, it just has to be used correctly.
Nice work, quite helpful and efficient
Thanks!
Might be a dumb question but would I be able to use the image and add items. So I’m interested in using an existing photo and simply adding rendered images. Would that be possible? Appreciate the advice.
You can also open the search function, f3 then type in quad view to see 4 views in one 3D view
That's a good idea :)
The CG Essentials I also created a new tab with the quad layout. So I can go there anytime
I just wish there was a “Dual” view
Thanks, very helpful. Been using Blender for months now, currently on 2.93
Awesome! It's a lot of fun :)
Cool ! You can actually trick the fpsy in addon and map a video version of the image.Maybe with people walking and interacting with the objects.
Hey Thankyou! It was very helpful!
Glad it helped!
this is super great, but what is the best way to match your units to the model so that you are modeling the exact size?
Awesome video! id love to see more on this like adding textures, interiors, also if its possible to use multiple photos from different sides as reference.thanks!
Thanks!
Hi, thanks for the video. Anyhow, I miss one information, the SIZE/SCALE. My feeling is that somewhere we have to manually define the size, but I don't see. where we are doing it.
I figured it out meanwhile. There is a "Reference distance" option on the left side. There you select an appropriate axis, where you have some known size information abou the object. A new scale appears on that axis. You move its two ends to the measured objects edges. Also you write the measured distance in the dialog box. If you save the fSpy file and import it to Blender, it will use proper scale.
Thank you brodie
id like to know how to add the reference image onto the model i guessing uv wrap must come into play just want to get it as realistic to photo as possible
Nice. Was just wondering. After importing the fspy data into blender and adding whatever additional models into the scene (after doing all the camera mapping and stuff); would it then be possible to transfer all that info to unreal engine to continue work? Is there a way to transfer the fspy camera info to an unreal camera? Not sure if I’m explaining properly. I ask u cos I know u don’t restrict yourself to blender alone.
Excellent! Thanks!
Glad it was helpful!
thank you your the best!!
Thank you, very helpful video.
Glad it was helpful!
Hi! can we do it for a 360 Equirectangular image or anyway to make that 360 into a 3d model?
Does anybody else have an issue where they make the fspy camera view and import it, but the image won’t load in?
The camera angle changes to the correct one, but I can’t import the image with it.
thanks!
how do you use the texture of the image to this barn?
Spy not supported in 3.x any alternative? Thanks
Could we import multiple Fspy files because the reference image has multiple angles?
Very handy!
Any projection mapping tutorials that takes off from this?
Awesome, man
Thanks!
This is indeed a very helpful tool. I don't know much about the Blender universe, so I don't know of alternative ways that might produce similar results. *However*, I would find it much more productive if the necessary step 1, the generation of the perspective would not take place in a stand alone tool, but directly in the Blender addon. Can this not be done somehow?
No idea - as far as I know, no one has done it, so it can't be that easy...
Can you then map onto each wall the relevant piece of photo and then if you move the 3D barn to have the wall with doors face on will the photo texture change perspective accordingly?
You can map the individual faces to different parts of the image, yes
thank you man you are awsome
Glad it helped!
And we’re just gonna leave a comment.
And we’re just gonna go ahead and like it.
oh nice addon :D
It's a good one :)
1:14
Which version of fspy should I download for windows
Fspy ia32 or fspy.win
Please help
i love it and like to see more videos
Ok thanks!
@@TheCGEssentials Thanks a lot for this tutorial as well. +1 for the idea for a video about using the textures from such perspective pictures
thank you
how did you added an edge loop at 9:05 ?
Press Control R.
Thank you.
You're welcome!
We will be happy if you make 2-3 more videos taking complex situations....
Well a lot depends on the level of complexity - it's best for projects that have very defined perspective lines - curved/non-standard shapes this might not be the right tool
It is shouting for missing libraries
Thank'ssssss broooo ^_^
No problem!
Very good ... +1
Thank you!
it doesn't work correctly. very few, ideal photographs are suitable for correct modeling with Fspy.
thx
dudes baked
Nope - just up super early making Blender videos so people can make fun of the way I look. You're welcome
Is there anyone here that woukd be willing to make cgi images from some photos for me to use for a movie project I’m working on?? I have lower quality pictures that I’d like to be made into cgi images so I can use them for cgi backgrounds for green screen work. If anyone is interested in doing this, how much would you charge me to do this??
why everyone is taking the fucking simple refrence for fspy😡😡😤
...what?
Funny how Blender Guru says when he started making videos he sounded terrible. "I sounded like this" then proceeds to sound exactly like Justin but just his starting part. That "Hey guys" part.
I'm not sure what exactly you're getting at with this...
@@TheCGEssentials If you sounded the same when you explain in a video as you do when you start a video I would have nothing to say. When you start a video it sounds too intimate, that voice like you don't want to wake up anyone in the house and whisper in my ear kinda voice. Hey! Maybe it just me so just ignore this comment.
@@dragonskunkstudio7582 It might just be you in this case...
@@dragonskunkstudio7582 me too
Come on, we need more complex house. This kind a house we dont need fspy