Відео

Thank you for your kind words, happy to help.

thank you, happy this helps.

Bill it's not spill but motion effect. When a subject is moving quickly in front of any green screen even at 60 fps there will be motion blur. That area of blur contains both the subject and the green background information making it difficult for chroma keying to get a good clean subtraction. This is true for any green screen. The solution is higher frame rates such as 120 fps which our next gen cameras support. Double the fps = half the exposure time which means you will need 2x the lighting on the subject.

welcome :)

This guy made a great video on how to fold up popup backgrounds. And yes it does take practice and patience. Once you get the hang of it - super easy. ua-cam.com/video/eHq9hT6l8vo/v-deo.html

If you have any questions - please drop me an email at info@sub2r.com

To transfer fusion nodes between clips: on the fusion page select the thumbnail of the new clip (the one you want to transfer nodes to) and then with the "middle-button-mouse" click on the clip with fusion nodes you want to transfer from. If there is no middle-button physically present on the mouse, you can copy and paste nodes between clips manually: highlight nodes from the clip you want to copy nodes from (excluding MediaIn and MediaOut nodes), then CTRL+C, select clip you want to copy nodes to (fusion tab) and paste in into the fusion composition area (CTRL+V). Reconnect MidiaIn and MidiaOut to newly pasted nodes. Neither left or right buttons will work with transfer nodes function in fusion, that is how Davinci Resolve programmed (at least for now, though users were asking developers for some kind of alternatives, but it has not been implemented). Note: at least on Color page, right click will have an option to "Apply Grade" so there is an alternative to middle-button-click on Color tab.

@@SUB2r : OMG! I'm thrilled! Vince will be the only character seen in each vide. Backgrounds change via blue screen. Base position during recording does not vary. So, that made things easier. I made screen capture of composition from earlier video and displayed in on a separate screen. Then copied all notes from earlier video, except MediaIn and MediaOut. Then I pasted copied content to the new raw video. Then connected the new MediaIn and MediaOut as shown in the previous version and POW! Looks fab! Tomorrow, the channel will post a new video "Be a Race Car Driver" using your method. Color etc looks great. Woohoo! Thanks again!

Hi Vince, there are two nodes (processes) that are slowing down a computer - CleanPlate and NoiseReduction. These two are the most compute intense and resource expensive. Turning them off (temporarily) will speed up real time playback. Once you are satisfied with the chroma substitution quality, you may disable these nodes and proceed working on remaining nodes until final render.

@@SUB2r : Ok. Will try again tomorrow afternoon. Had about 99% of images masked and there were a few minor holes. Couldn’t recall setting to “fill holes” and made it worse. Underlying problem is light glare from nearby white wall. Covered wall with black fabric to defuse glare and adjusted light position. Will try both and use ur process after recording.

@@VinceLatte Also when I used to process files on older PC, I would pre-denoise recordings first and then drop re-rendered denoised files into chroma project. Even though it would be a two-step process, it saved me a lot of time during chroma work (faster real time render). There is one more trick when working with large format files (2-4k resolution) is to Optimize media in Davinci Resolve, it would create optimized files (takes hard drive space) to speed up Davinci internal processes. May have a separate tutorial on how to work large files in DR with optimized media.

@@SUB2r The coat, suit, and hat worn in my videos are made of a fuzzy fabric and are bright white. I chose to use a dark blue screen (cloth) for chroma key instead of green or black. Green bleeds into the white fabric, and since my beard is black, a black chroma screen wouldn't work either. This choice allows for better color separation. However, the shine in the eyes causes a glare, possibly worsened by a white wall off-camera. Previously, I relied on Magic Mask. I'd do something like, take the raw video, use a blue marker to outline the body (Vince), and render it left/right in one click. After rendering, I'd go to the Color tab, right click to add Alpha output, connect the blue pipe from the source to the standalone blue dot, and then switch back to the Edit tab. Easy-peasy! However, after a recent app update, I noticed a change in Magic Mask. When I follow the same process with the blue marker inside the desired image, rendering in the Color tab shows angular lines inside and outside the image. I end up spending hours adjusting the blue line and adding red lines to correct it, which is frustrating! Recently I was 99% done with video image when I noticed some holes. It was the glare from the eyes. I tried to fix the holes and and too impatient at the time to rewatch video to find fill feature. I tried Magic Mask AGAIN... :-( to only make it worse. So close... To address the glare, I've added black fabric to cover the white wall and slightly adjusted the lighting. I'm hopeful this will make a significant difference in reducing the glare on that side of the frame. Will try your process again with recording changes.

@@VinceLatte If I am understanding the issue, eyes reflect the blue color bouncing off the environment and it makes eyes disappear after chroma substitution? If this is the case, a simple shape mask (polygon) can be introduced over face/eyes and connected to the DeltaKeyer node blue triangle (make sure to select "Invert" in polygon controls). This will exclude everything inside the polygon from DeltaKeyer chroma removal. If head moves around, then simple automation can be done to polygon to follow the movement.

The good news is there are several ways to do this in Davinci. Macro is probably most difficult and time-consuming process which is great to share nodes chain between different users but may be an overkill for your needs. The simplest way is to Save As existing complete project as new project (assuming it has at least one clip that has desired fusion node structure), you can delete all but a clip that you want to transfer nodes from to make workspace cleaner, then load a new clip into timeline, then go to Fusion page, make sure that “Clips” icon is active at the left-top menu bar (this will make all clips in the timeline down below visible), then select the clip you want to apply fusion nodes to and then with mouse middle button click on the clip you want to transfer nodes from. Agree to overwrite existing fusion nodes (pop-up screen) and that’s that. Similar process can be done in color page to transfer color nodes between clips.

Thank you Serge; I appreciate your advice more than you know. While DaVinci Resolve is a wonderful NLE, I know it will take a long time to master. But learning from your experience is very helpful. Jay🙏🏼

@@jaybcreighton happy to help, Jay!

It is a TUTORIAL about how to use our product and Davinci Resolve. So yes ... by definition we talk about our product.

All requests for eval units need to go to info@sub2r.com Thanks

Larger versions are on the road map.

You are welcome!

Thank you for the comment. In most cases this level of editing is not required. This particular tutorial was made for anyone who wants best possible outcome and needs white background, which in my opinion is the most challenging. Will be creating more tutorials in the near future using different methods for both speedy and/or quality output.

@@SUB2r great!! That would be awesome. I'm use to keying in FCP and find Resolve harder to key. So I hope you do more tutorials!!

yes, will definitely be making more tutorials to explain what and how was it done.

How would you suggest the lighting be setup?

@@SUB2r I wholeheartedly disagree with the above comment. Given the presented use case as an OBS livestream, your lighting is exceptional compared to what the average person with a pop up green screen can realistically ascertain--particularly in such a confined space. I've got my eyes set on this myself and more than a few ideas about how to use it in some creative livestreaming applications!

I'm sure other sizes are a great idea

Our patented technology is scalable and larger versions are on our road map - we need to get this model successfully in the market first. While not recommended - we were surprised how well the MacGyvered setup worked.