Getting started with RealityCapture with Timothy Hanson - webinar recording
ฝัง
- เผยแพร่เมื่อ 2 ส.ค. 2024
- In this webinar, Timothy Hanson will demonstrate how to take small to medium size prop and with the use of RealityCapture, turn it into a photorealistic digital asset for use in VFX.
The webinar will cover:
- Shooting tips on how to properly capture an object
- Basics of the User Interface
- How to work with components and use Control Points
- How to clean model in RealityCapture and filter unwanted parts
- Post-Processing in Zbrush
Don't forget to use code provided by Timothy Hanson "RC30" which will give 30% off all his tutorials at maxdepth.tv/. Check it out!
Also if you wish to follow along with the webinar, here is a licensed dataset of the small statue used in the video which you can process in RC and export for free: www.capturingreality.com/down...
Download RealityCapture (PPI) for FREE here: bit.ly/2sotr7h
Try it yourself with free sample datasets: bit.ly/3cRBfQp
Follow us on:
Instagram / capturingrealitycom
LinkedIn / capturing-reality
Facebook / capturingrealitycom
Twitter / realitycapture_ - วิทยาศาสตร์และเทคโนโลยี
![เจ้าผู้เดียว - ดิด คิตตี้ [OFFICIAL MV]](http://i.ytimg.com/vi/q0NLCYKBOTo/mqdefault.jpg)
Great webinar! Tim was very nice and helpful and he answered all the questions. Thank you RC and hope in the future we will see other creators sharing.
Great webinar. Very comprehensive, and helpful to learn CR more quickly.
One thing i missed in the tutorial: I always run Preview reconstruction (which is superfast and includes coloring) to check for errors in mesh and to make the reconstruction Selection more precise... Helps a lot!
Thank you for creating so much value for the community! A transcript mode to go through the video faster would be quite nice . Have a look at tldv or otter :) When will be the next session? Best, Rolf
Is there a general rule for best focal length to use for taking photos for reality capture? Sorry if I missed this in the video.
good contents, my friend
It would be a lot easier to have strong light inside towards the object and object on turning table. Then you could go even F22 with low iso and would not need to use photo editing to get it brighter. Also your DSLR could be mounted stable and the object could be turned instead of moving the camera... This way every side you take photo of will be bright without shadows and enough light, as well as focus will be exactly same.
What camera is he using? and what is the settings for an iphone if you dont have a camera like his using.
Your better off using a higher iso than brightening in post. The better solution is to lower the shutter speed to achieve proper exposure.
what if you wanted the details of underneath the feet? How would you ensure that the object is complete captured?
Hi, then you can use image masks: th-cam.com/video/xxsdbzBwWLE/w-d-xo.html
👍
Not sure 28mm on the drone equates to 28mm on a FF DSLR? There would be a crop factor from the smaller sensor on the drone. For example a 35mm lens on a canon 5D is 35mm, but a 35mm lens on a m43 sensor camera like a Panasonic gh5 (which has a crop factor of 2x), the field of view on that setup has the same field of view as a canon 5D with a 70mm lens on it (2x 35mm).
So you’d want to work out the crop factor of your drone vs it’s focal length, and then match that on your dslr.
just when you were about to export from zbrush and re-import in RC you cutted the video?? The part I was looking for... :(
Is shooting RAW photo's recommended? Or would normal work qualitatively well too?
RAW is used in this tutorial to maximize the amount of details you can get - optimized and converted to TIF in Lightroom.
Direct in-camrea JPEGs work too, but you might lose little bit of image data in sub-optimal lighting scenarios.
Can you please go over using the mavic 2 pro as it is one of the most widely used drones
There is new series Drone mapping guide: th-cam.com/video/j62Xgb2EYNg/w-d-xo.html
47:00 confused why you'd need to export then re import the components, shouldn't they already be in the project if you're looking at them? You can't control point them before reimporting? Unless it's just convoluted programming or am I missing something.
It was just an example how to solve if you have more components after alignment.
우정의종각!
Thank for this Video. I have a question. It is possible to place an object on a turntable and shot photos? What i mean is the object is rotating, and the Camera is on one point.
Yes, it is possible. And maybe better than it this video. It depends on the size of the object
@@CapturingReality The Objects are not so big. We have a professional turntable cabin with 3 cameras. The Photos look great, but i have problems to combine all cameras in CaptureReality Software.
@@dariuswojcik3256 How looks your setup? Is there something in the background? What is the overlap between cameras?
@@CapturingReality Many thanks for your interest. I have two packages of Pictures 3x72 Pictures (0° from the site, 45° and 90° from the top) From the Top I use only one picture. Two packages because one stay the Car Toy on the wheels and one i rotate the Car on the roof. The background is white. The Problem is CaptureReality make 1 component for each Cameras. I make for any component a mask, but it doesn't work to make a one big component. I hope you understand me.
@@dariuswojcik3256 So it seems, that you need to use control points to merge these components together. Or you can try to create more angles, not just three.
super
Would it work if I take photos from the same spot only changing angles and rotate an object? It could be done with white background and softbox, would it work with dslr?
Yes, it would work. For this are using some rotation tables. But as you said, you need to have a background without texture. Yes, you can use dslr camera.
@@CapturingReality, I'm experementing with shoe modeling, planning full custom carbon road shoe, now I'm scanning ordinary shoe, then copy of my leg in plaster, and when I have to make an adapter from an plaster leg to a cycling cleat, which has a specific curve
How do you import the photos I'm a newbie any suggestions on fresh beginner tutorials?
Hi, you can just drag and drop the images or import them by Workflow tab/Add imagery/Inputs or Folder.
@@CapturingReality thanks for responding to my question do guys have a support team for beginners?
@@Mikey.Henderson We have a support, but there are mostly solving problems with applications, licensing... For beginners we have video tutorials on TH-cam and Forum, where you can find a lot of answers for your questions.
Would a macro lens work ok
They could work, but it depends what are you capturing and how.
wait...How did you select faces at 1:00:57 ?
At 1:00:27 you can see him activating a Lasso selection tool, which can be found in Reconstruction tab. This tool can be used to freely select mesh parts.
@@kurkomat7220 Thanks!! i just wanted to know how did he select face mode during lasso selection.
If you use a lazy Susan base and rotate the base keeping camera steady on tripod produce better results? Even if not, it may be easier to do for multiple objects.. just a thought. Great insight on video.. awesome 👌
plenty of places to buy a lazy Susan! LOL
Great stuff, i know you pondered upon handling glass/windows while shooting a car (for ex), some tips to photograph lot of glass material, would help. thanks.
Hi, we are planning to create tutorials regarding this issue. For example this is an older one: th-cam.com/video/D7iY146cU2Y/w-d-xo.html
@@CapturingReality Thanks, appreciate the response.
how many images are too much images ?
It depends on various things. The size of the model, the images overlap, your PC parameters, the way how you are capturing, etc.
It would be cool if you got a machine where you can place the object in and the machine automatically takes a lot of pictures and exports it to Blender
I’m thinking about buying/making a slowly rotating platform to place my objects on. Camera on a tripod set to take a photo every 1-2 seconds. Should be a bit quicker.
Great this tut is Fucking greate. And it is simple as F :). First tut that I did not want to destroy my PC xD.
lol my reality capture does not look like this... -___-....
like his 1Ds view window is showing "994/996 cams" lol where do i find THAT data.... -___-
@@torodatruth Hi, what do you see? Do you have set the layout to for example 1+1+1? If so, then click and hold on 2Ds view button and you will see the 1Ds option. Then just click on it and you will have opened 1Ds view.
13:45 "...if we go here and we take this f-stop and take it lower - if you know anything photography - we're going to get a lot more light into the camera but we're also going to be introducing a lot more depth of field". That is not true is it - you are reducing the depth of field surely?
I think that such a colorful background table is totally wrong
Sorry Tim, you are completely wrong on f NUMBERS!
Thank you for creating so much value for the community! A transcript mode to go through the video faster would be quite nice . Have a look at tldv or otter :) When will be the next session? Best, Rolf