Re: [Cin] Cool stuff
Both interesting information. Thanks for pointing them out. GG always laughs about "Super Resolution" because the human eye can not really see it anyway. Phyllis On Sun, Oct 4, 2020 at 5:56 PM Andrew Randrianasulu via Cin < cin@lists.cinelerra-gg.org> wrote: > В сообщении от Wednesday 16 September 2020 18:26:07 Sam via Cin написал(а): > > Hi guys, > > > > i have found an interesting video which shows how you can convert > > footage with a lower FPS to a video with a much higher FPS without any > > losses. > > Not exactly this type of cools stuff, but I found interesting article: > > > https://www.collabora.com/news-and-blog/blog/2020/09/21/open-source-meets-super-resolution-part-1/ > > quote > Performance > Despite their great upscaling performance, deep learning backed > Super-Resolution methods cannot be easily applied to real-world > applications due to their heavy computational requirements. At Collabora we > have addressed this issue by introducing an accurate and light-weight deep > network for video super-resolution. > > To achieve a good tradeoff between computational complexity and > reproduction quality, we implemented a cascading mechanism on top of a > standard network architecture, producing a light-weight solution. We also > used a multi-tile approach in which we divide a large input into smaller > tiles to better utilize memory bandwidth and overcome size constraints > posed by certain frameworks and devices. Multi-tile significantly improves > inference speed. This approach can be extended from single image SR to > video SR where video frames are treated as a group of multiple tiles. > > We designed our solution on top of the open-source Panfrost video driver, > allowing us to offload compute to the GPU. > > Coming up in Part 2 of this series, we'll take a deep dive into how our > model works, and how you can use free, open source software to achieve a > higher level of compression than existing video compression methods. Stay > tuned! > ---quote end > > Might be interesting read, waiting for second part! > > PS: as far as I understand this "AI" thing actually just 'invent' a lot of > specialized algorithms on the fly (while training) and then apply them if > similar situation in image occurs ? > > > > > > > https://youtu.be/sFN9dzw0qH8 > > > > https://github.com/baowenbo/DAIN > > > > The nice thing is that this software is free of charge. For my > > commercial projects I mainly use Da Vinci Resolve and there is the same > > possibility which I use again and again. This feature is great if you > > want to create slow motion effects and you only have footage that was > > created with a low FPS. I don't know if one day this software will be > > used for Cinelerra, but it would be a great feature to use. Maybe you > > can transfer footage from Cinelerra to this tool with a mouse click and > > transfer it back to Cinelerra after the rendering is finished? Even if > > an integration would not be possible, you can use this tool > independently. > > > > Sam > > > > P.S.: By the way, the problems with the language plugin and web site are > > still bothering me, but a solution will surely be found soon. > > > > > -- > Cin mailing list > Cin@lists.cinelerra-gg.org > https://lists.cinelerra-gg.org/mailman/listinfo/cin > -- Cin mailing list Cin@lists.cinelerra-gg.org https://lists.cinelerra-gg.org/mailman/listinfo/cin
Re: [Cin] Cool stuff
В сообщении от Wednesday 16 September 2020 18:26:07 Sam via Cin написал(а): > Hi guys, > > i have found an interesting video which shows how you can convert > footage with a lower FPS to a video with a much higher FPS without any > losses. Not exactly this type of cools stuff, but I found interesting article: https://www.collabora.com/news-and-blog/blog/2020/09/21/open-source-meets-super-resolution-part-1/ quote Performance Despite their great upscaling performance, deep learning backed Super-Resolution methods cannot be easily applied to real-world applications due to their heavy computational requirements. At Collabora we have addressed this issue by introducing an accurate and light-weight deep network for video super-resolution. To achieve a good tradeoff between computational complexity and reproduction quality, we implemented a cascading mechanism on top of a standard network architecture, producing a light-weight solution. We also used a multi-tile approach in which we divide a large input into smaller tiles to better utilize memory bandwidth and overcome size constraints posed by certain frameworks and devices. Multi-tile significantly improves inference speed. This approach can be extended from single image SR to video SR where video frames are treated as a group of multiple tiles. We designed our solution on top of the open-source Panfrost video driver, allowing us to offload compute to the GPU. Coming up in Part 2 of this series, we'll take a deep dive into how our model works, and how you can use free, open source software to achieve a higher level of compression than existing video compression methods. Stay tuned! ---quote end Might be interesting read, waiting for second part! PS: as far as I understand this "AI" thing actually just 'invent' a lot of specialized algorithms on the fly (while training) and then apply them if similar situation in image occurs ? > > https://youtu.be/sFN9dzw0qH8 > > https://github.com/baowenbo/DAIN > > The nice thing is that this software is free of charge. For my > commercial projects I mainly use Da Vinci Resolve and there is the same > possibility which I use again and again. This feature is great if you > want to create slow motion effects and you only have footage that was > created with a low FPS. I don't know if one day this software will be > used for Cinelerra, but it would be a great feature to use. Maybe you > can transfer footage from Cinelerra to this tool with a mouse click and > transfer it back to Cinelerra after the rendering is finished? Even if > an integration would not be possible, you can use this tool independently. > > Sam > > P.S.: By the way, the problems with the language plugin and web site are > still bothering me, but a solution will surely be found soon. > -- Cin mailing list Cin@lists.cinelerra-gg.org https://lists.cinelerra-gg.org/mailman/listinfo/cin
Re: [Cin] Cool stuff
В сообщении от Wednesday 16 September 2020 18:26:07 Sam via Cin написал(а): > Hi guys, > > i have found an interesting video which shows how you can convert > footage with a lower FPS to a video with a much higher FPS without any > losses. > > https://youtu.be/sFN9dzw0qH8 > > https://github.com/baowenbo/DAIN --- Requirements and Dependencies Ubuntu (We test with Ubuntu = 16.04.5 LTS) Python (We test with Python = 3.6.8 in Anaconda3 = 4.1.1) Cuda & Cudnn (We test with Cuda = 9.0 and Cudnn = 7.0) PyTorch (The customized depth-aware flow projection and other layers require ATen API in PyTorch = 1.0.0) GCC (Compiling PyTorch 1.0.0 extension files (.c/.cu) requires gcc = 4.9.1 and nvcc = 9.0 compilers) NVIDIA GPU (We use Titan X (Pascal) with compute = 6.1, but we support compute_50/52/60/61 devices, should you have devices with higher compute capability, please revise this) quite a lot of Nvidia dependencies :} But may be someone will port it to OpenCL ...someday. Thanks anyway! > > The nice thing is that this software is free of charge. For my > commercial projects I mainly use Da Vinci Resolve and there is the same > possibility which I use again and again. This feature is great if you > want to create slow motion effects and you only have footage that was > created with a low FPS. I don't know if one day this software will be > used for Cinelerra, but it would be a great feature to use. Maybe you > can transfer footage from Cinelerra to this tool with a mouse click and > transfer it back to Cinelerra after the rendering is finished? Even if > an integration would not be possible, you can use this tool independently. > > Sam > > P.S.: By the way, the problems with the language plugin and web site are > still bothering me, but a solution will surely be found soon. > -- Cin mailing list Cin@lists.cinelerra-gg.org https://lists.cinelerra-gg.org/mailman/listinfo/cin
[Cin] Cool stuff
Hi guys, i have found an interesting video which shows how you can convert footage with a lower FPS to a video with a much higher FPS without any losses. https://youtu.be/sFN9dzw0qH8 https://github.com/baowenbo/DAIN The nice thing is that this software is free of charge. For my commercial projects I mainly use Da Vinci Resolve and there is the same possibility which I use again and again. This feature is great if you want to create slow motion effects and you only have footage that was created with a low FPS. I don't know if one day this software will be used for Cinelerra, but it would be a great feature to use. Maybe you can transfer footage from Cinelerra to this tool with a mouse click and transfer it back to Cinelerra after the rendering is finished? Even if an integration would not be possible, you can use this tool independently. Sam P.S.: By the way, the problems with the language plugin and web site are still bothering me, but a solution will surely be found soon. -- Cin mailing list Cin@lists.cinelerra-gg.org https://lists.cinelerra-gg.org/mailman/listinfo/cin