Coder Social home page Coder Social logo

il_video_inpainting's People

Contributors

haotianz94 avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar

il_video_inpainting's Issues

Code issue

I want to ask whether the PID repair algorithm and the video repair algorithm are run separately in the code, whether the PID is used to repair every picture first and then the whole picture is repaired, or whether the image repair algorithm and the video repair are carried out simultaneously?

Running time is a little long

For example:

python3 train.py --train_mode DIP-Vid-Flow --video_path data/bmx-trees.avi --mask_path data/bmx-trees_mask.avi --resize 192 384 --res_dir result/DIP_Vid_Flow

It needs more than 6 hours on a single 1080ti.

Reproducibility of the results

Hello, I've run your model on the video of 32 frames and got PSNR of 28 instead of 32.

How exactly was PSNR calculated? How many frames in the video were considered for testing?

Training and Testing paradigm

Hi,

  1. (EDIT: This query is clear to me now. ) Could you, in simple words, kindly explain the training and testing parts of the proposed methodology?

  2. How do you generate the whole impainted video in the end? I see that that the generator produces a single image at a time.

Pardon me if the questions are naive, but I just wanted to be sure of my understanding. Also, thank you for making your implementation available!

Results on Davis dataset.

Thanks to the authors for sharing the code. Do you have any plan on releasing pre-computed results on the Davis dataset so that we can make a direct comparison to your results in a follow-up paper? Many thanks!

Calculating PSNR to compare results

Hello authors, thank you for your great work!

I am a researcher working on video inpainting. I would like to compare my work with yours quantitatively. Would it be possible for you to provide code or instructions as to evaluating your method? In particular, could you provide what is necessary to reproduce the PSNR numbers in your paper?

Best,
Luke

How to get the inpainted video?

After training, I got 20 folders which are corresponding to the 20 passed while training. And I found the images in each folder, but how can I get the inpainted video? I tried to use the images in folder "result\DIP_Vid_Flow\bmx-trees\020\final\full" to make videos. But the effect was bad. So I need your help.

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.