Quote:
With a standard supervised neural network system you need to have many training samples of TBCed vs non-TBCed clips, which are quite difficult to obtain, especially for a very wide range of types of videos (cartoon, live-action, high detail, low detail, etc.) and jitter (smooth wobble, jagged wobble, etc.) that a practical set of software TBC tools need to cover. The other alternatives are unsupervised neural networks and "untrained" neural networks (which is an unsupervised neural network being trained on-the-spot). Unsupervised neural networks may be more robust to many examples but aren't going to perform as well in general. Untrained neural networks can visually output superior results but are very very memory- and time-hungry. I will consider exploring neural networks for TBCs, but I don't expect them to be the de facto TBCs for the reasons mentioned above. |
Hi did see this about paper-shredder machines how to reconstruction of strip-shredded text documents i was thinking if there is some help or tips how to do software TBC :)
http://sibgrapi.sid.inpe.br/col/sid....%20ID%2088.pdf and https://github.com/thiagopx/deeprec-sib18 are you using OpenCV library in python for this ? i have tested it a little in c++ language but im not a programmer vhs maybe to bad quality to use Edge Detection like this https://medium.com/sicara/opencv-edg...l-7c3303f10788 to align the fields just thinking out loud :) |
Quote:
For Spatial TBC 1 and Temporal TBC 1 I'm just using PyTorch. I'm going to explore a few more ideas, some of which will use stuff from OpenCV. Quote:
|
ok interesting to see how it progress :)
|
2 Attachment(s)
imgkd any progress on the TBC software :)
i myself have been testing jmac698 TBC V0.61 and trying to understand how it works if i understand it use a gradient mask to detect better the left and right side edges like this Picture i Attached it and then use minmax() function to detect the lowest pixel value at every line here is that function POST Nr:3 https://forum.doom9.org/showthread.php?t=162790 and with that it creates a clip that is 2 pixel wide and that have the inputs clips hight 2 pixel wide because Left half side lowest pixel value and Right half side lowest pixel value and that creats the offset clip that goes then in to the Rescale() function and that use the dejitter() function to offset the line by the lowest pixel value per line example if the luma value is 3 it shift it 3 pixels then to the right or left but this is my best guess about jmac698 TBC how it works i did use RemoveGrain(11) on the clip i did input to the findpos_h() function that creats the offset clip and that did help little i was thinking if it's better to resize the video only vertical to get something to work better in the tbc software i did test this function to RemoveDeadPixels(x, 1, 7, 255) to change the lowest pixel to white X in the parameter is the horizontal coordinate to the pixel to change to white and then use a align function to align more info here about RemoveDeadPixels function POST Nr:11 https://forum.doom9.org/showthread.p...915#post699915 a good old app to pixel peeping i use is Mouse zoom 1.5 |
I've written down a couple new ideas to try out, and have been developing one idea in particular, but that's all. No new TBCs to show for now, though.
I've mostly been waiting for LS to give me some more samples so I can test out these TBCs on a wider variety of content. Given the sheer novelty of this stuff (as something that actually works), I have been entertaining the thought of developing this into proprietary software. If I start being less specific in future posts, this is why. jmac's software TBC project is mainly considered a failure outside of a few toy samples*. The main problem is that if you're producing a piece of software that relies purely on the image content, such as software TBC, it is by definition a computer vision problem. You must treat it like a computer vision problem. Jmac didn't treat it like one. (He proposed later ideas that were closer, but he never actually prototyped them.) One may argue that jmac's TBC relies on the black borders. But what are "black borders"? In real-world cases they're almost never black, the borders are not always reliable, and regardless it's still part of the image content. And often times you don't have the borders at all (a good software TBC shouldn't need the borders). Computer vision is a major (and very broad) field consisting of tons of tools, various algorithms, and various high-level methodologies. So to write a good software TBC (or in reality, multiple software TBCs), you need to study lots of techniques in the computer vision field and have a basic understanding of how they work, their pros and cons and what they're best for. Followed by lots of trial and error. This is what makes my suite of software TBCs distinct from Jmac's prototype. -------- Also, I believe Jmac details this algorithm in the first post of the filter: Quote:
Quote:
-------- *To contrast this, the only thing I'm using to test my software TBCs are real-world examples. Therefore, if the TBCs work, they work on real-world samples. |
ok thanks for the update and information :)
|
Any progress on this? this looks like something I could greatly use as I am having a hard time finding a Field TBC tht isnt an arm and a leg in price, and my VCR only has a line tbc.
As a dev, I think python is the better choice, especially with all the integrations you can do in the future with AI/ML. Even if we dont train our own, existing stuff can just improve image quality overall. I have seen some crazy AI enhancement of very old footage. Youtube has a lot of examples like https://www.youtube.com/watch?v=EjVzjxihGvU AS for proprietary. I encourage keeping it open source and allowing the community to work on, but I know everyone deserves to make money how they can. I just hope you keep the pricing reasonable if you do monetize. |
Quote:
Quote:
Quote:
|
Quote:
And I know you say many times it can be purchased and then sold, but you still need to fork over the money, and life gets in the way and 2-3 years later, its still sitting in the closet. So it's not a great option for many people, because that's just how people work. I posted here a couple years ago. Got a VS30U, and the ATI Wonder 600 USB. But an external TBC has always alluded me due to cost (I managed to get a KeyWest Voodoo for $70, but it caused audio sync issues and died about 20 mins in). I got the VCR for like $50 working, and the 600 USB for something similar if not cheaper. But then the TBC prices re $200+ and rising. Just a balance of quality and money. So if we can use software to augment that, it could save people money. I don't expect it to be as good as external, but even if its 30-50% as good, it's better than 0. And even just that could be good enough for some to ignore the external TBC. Because at this point, I am about to just give up and capture VCR > PC directly, and call it a day and just hope I get decent quality without an external TBC Unfortunately I am not great at AI or Video, even if I am a dev, I specialized in a different direction for my career, so it would take me good while to ramp up. I would need to dig deeper than my surface level understanding of all the technical aspects of analog video capture. So I get it LordSmurf, and I know you are right and where you are coming from, and I respect your experience, but as hardware becomes increasingly harder and harder to get, some of this stuff needs to move to software, or we need to start designing an open source board to do all this. Maybe push into the maker scene and program a bunch of arduinos to do what 1 board does, or design a custom board. Either way, we can't just rely on old hardware forever. We might need to program an arduino to do what 1 tiny chip did on the original board. But we have to start trying stuff and pushing the envelope (i wish i was more into hardware too at this point). |
Quote:
Quote:
- If hobby -- All hobbies have costs (DIY video = hobby, even if short-term hobby). Buy it, use it, resell it. - If business -- All businesses require capex. Quit being a cheapskate, it will only result in horrible quality that should never be foisted upon paying customers of services. Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
|
After reading up on how a frame TBC worked, I was like "You can do that in code" but something clicked finally about analog signals. I was completely forgetting the part that in order for any software to process that signal, it needs to be converted to a digital signal. and the Analog to digital conversion is how you lose the data needed to create a frame. Now I realize where i was making an error in my logic.
I also have kind of derailed this topic, though I do want to add a last thing. Now that I realize what you are getting at, maybe the key isnt a TBC per say, but the Analog to Digital conversion. We do various conversions to an image. But maybe we need to take the analog signal and convert it to just labeled data? take each line of video, or whatever we can get, label it up with all the meta data, and send it on. Then recreate images from the data + labels and then use the label information to do things normally you couldn't in software? Just spitballing, I know its probably been considered already, but I feel like there is something there. Thank you for the ES15/DVK suggestion. I might go this route, because even if it not super cheap, its multiple parts that can be purchased over time and better budgeted for. And you mentioned hobby or business, but im not really either. THis is a 1 time project for me. The only reason i am going this far, is these tapes are the only remnant of memories of my wife's father, who she lost in her early 20s. Due to fires, and time , pictures, and many things are lost. One of those is voice, which is only on these tapes. So I want to do everything I can to squeeze as much as possible out, without breaking the bank on top tier equipment. |
Quote:
This is not likely to be an "easy button" newbie/dummy-friendly approach, but I imagine it to be more like barrel/pincushion distortion correction in camera lenses. Yes, it will have a set of profiles for common situations, but it will have some manual correction ability as well. Quote:
Arguably, line TBC should have been inclusive as well. All DVD recorders (not just the rare ES10/15), all capture cards. But that's just not what happened. So what we have now are lots of wiggly crappy videos that need to be saved with a software TBC -- if it can exist. And I think we're getting close now, with IMGKD leading the development. Quote:
Quote:
... and I think enough derail for now. :wink2: |
Year old thread but this is my exact question.
They say RF capture allows you to bypass the need for a frame TBC entirely. When you are referring to software TBC here, I'm assuming you mean line correction and not frame as pointed out above. However RF capture briefly spoken about in this thread as being (not useful), some have stated it allows you to bypass the need for frame correction or frame TBCs. Is this true, and why would that be? |
Yes it's true check your thread. A TBC is a just a capture device that doesn't convert back to analog, If you have a way to get that perfectly timed digital signal straight to computer wouldn't you want that?
|
Quote:
Continue this topic here: http://www.digitalfaq.com/forum/vide...quired-rf.html Quote:
- Timing correction (TBC) is required for consumer analog formats, especially VHS. - To date, no software TBC exists, referring to Win/Mac/Linux type software, what most people think of as "software" (or apps, etc). No such TBC may ever exist. - Dedicated chips are currently needed for TBC, and may always be the only way. And no, the on-chip software/firmware cannot simply be ported. This thread is for a project that stalled out. Both time, and because software TBC is an uphill battle. |
Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.