Home Artists Posts Import Register

Content

  • So exactly, what is Dain-App?

Dain-App is a windows interface that manage movies, the frames are sent to the DAIN Backbone where all the magic happen. So pretty much is just a movie handler with a easy interface.

  • So, do i need to pay something to use the app?

No, the app is free, patrons only get access a week earlier to the build a way to thank the support for the project.

  • Ok, so what is happening to my donated money?

Life is juggling a lot of projects in the air to see what help you pay the bills, with donating is easier to focus on this project to complete it faster.

This project is made with the collaboration of Wenbo Bao, the creator of DAIN, from the beginning we had a agreement to share the profits 50%-50% with is fair, without his support this app would not exist.

If the project go really well, i might also get a Nvidia Graphic card, right now i test what i can in my computer and the real interpolation in google colab, since my computer don't support CUDA.

  • So, what to expect from this project in the future?

There is quite a few ideas (will do another post about those later on) that i want to implement, after that the project may be complete, but only the future will tell.

  • I am a developer and want to train my own models. Can it work with Dain-App?

Sure, just replace the model_weights/best.pth file.

  • I am getting an OutOfMemory error?

The app use a lot of memory, you can resize the resolution of the video to fix that, or from version >= 0.3 you can split the frame in smaller areas to fix it.


I will try to keep this post updated with new information and updates.


Comments

Anonymous

Hello! I have a question about CUDA. To use your app I have to download the CUDA toolkit from nvidia website, correct? Or is there something else I should do?

DAINAPP

I'm not totally sure that the toolkit is needed, just update your driver graphic card if needed. But your graphic card also need to be CUDA compatible.

Anonymous

Hi, it seems there's a problem with this software on my PC. When I start render it just freezes and after a minute it just crashes and that's it.

Anonymous

Log of the console: cudnn is used OK C:/Gargoyles.mp4 Input FPS: 23.976023976023978 C:/Output C:/Gargoyles.mp4 C:/Output/Gargoyles//Gargoyles.mp4 23.976023976023978 0 0 0

Anonymous

How can I get DAIN to run with pytorch for CUDA 9.x?

DAINAPP

Are you taking about the git project? Dain-app don't need to setup anything

Anonymous

Hello. This problem has occurred recently when using your software. How can I solve it?

Anonymous

App Crash Log Traceback (most recent call last): File "my_design.py", line 131, in render File "my_DAIN_class.py", line 779, in RenderVideo File "my_DAIN_class.py", line 524, in Configure File "networks\DAIN.py", line 48, in __init__ File "PWCNet\PWCNet.py", line 543, in pwc_dc_net File "PWCNet\PWCNet.py", line 144, in __init__ File "site-packages\torch\cuda\__init__.py", line 194, in _lazy_init File "site-packages\torch\cuda\__init__.py", line 80, in _load_cudart File "site-packages\torch\cuda\__init__.py", line 53, in find_cuda_windows_lib UnicodeDecodeError: 'utf-8' codec can't decode byte 0xb2 in position 3: invalid start byte

Anonymous

Hi, I just run the DAIN and there's a problem with the output. After DAIN says finished, there's nothing inside the folder that I choose. So may I ask how can I fix that problem?

DAINAPP

Hey there, ca you send me a print of your application and console? I think is easier to talk to me in discord. I'm "GRisk"

Anonymous

Hi, i've downloaded the latest 0.31 version and got this error: App Crash Log Traceback (most recent call last): File "my_design.py", line 131, in render File "my_DAIN_class.py", line 779, in RenderVideo File "my_DAIN_class.py", line 524, in Configure File "networks\DAIN.py", line 48, in __init__ File "PWCNet\PWCNet.py", line 543, in pwc_dc_net File "PWCNet\PWCNet.py", line 144, in __init__ File "site-packages\torch\cuda\__init__.py", line 194, in _lazy_init File "site-packages\torch\cuda\__init__.py", line 80, in _load_cudart File "site-packages\torch\cuda\__init__.py", line 53, in find_cuda_windows_lib UnicodeDecodeError: 'utf-8' codec can't decode byte 0x80 in position 9: invalid start byte

Anonymous

CUDA out of memory. Tried to allocate 192.00 MiB (GPU 0; 2.00 GiB total capacity; 1.01 GiB already allocated; 186.23 MiB free; 162.76 MiB cached) So, I guess my GPU is not appropriate. Sad...

Anonymous

Every time I try to render using the PNG Sequence, the command center gives me an AssertionError message with the Error prompt appearing as Crash [Saved on crash_log.txt]: Check out Discord for possible solutions: Check Errors fix on Discord. Error:

Anonymous

I have a question about What spec of GPU is the best for dain app?

Anonymous

I don't understand how I'm supposed to run this. It just opens a blank command prompt for one second and then closes. It does nothing. There are no installation instructions anywhere so I have no idea wtf I'm supposed to do.

Anonymous

I'm in the same situation. I was hoping to find a post explaining how to use it but if there is one I've not found it yet.

Sergiy

Dear DAIN-APP developer. I'm very impressed by the capabilities of your app. In one of your YouTube video "Depth-Aware Video Frame Interpolation" you shows "Estimated depth Map" stream. Is there is a way to output "Estimated depth Map" stream sequence as well (preferably gray-scale)?