{ "id": 121313, "title": "GAN and DeepFake resources", "url": "/competitions/deepfake-detection-challenge/discussion/121313", "author_name": "Gabriel Preda", "post_date": "2019-12-12T11:03:57.719000", "votes": 95, "comment_count": 29, "views": 0, "content": "

This resources list is not exhaustive, it provides just a starting point for Kagglers that would like to join this competition in order to learn, like myself.

\n

I provide here technical articles links, small blog articles, links to Github projects and some inspirational Kaggle Kernels about DeepFake, DCGANs and GANs.

\n

\"Wired:

\n

Blogs, Technical Articles

\n\n

Tutorials

\n\n

Kaggle Kernels

\n\n

Github repos

\n", "messages": [ { "id": 693424, "postDate": "2019-12-12T11:03:57.720Z", "content": "

This resources list is not exhaustive, it provides just a starting point for Kagglers that would like to join this competition in order to learn, like myself.

\n

I provide here technical articles links, small blog articles, links to Github projects and some inspirational Kaggle Kernels about DeepFake, DCGANs and GANs.

\n

\"Wired:

\n

Blogs, Technical Articles

\n\n

Tutorials

\n\n

Kaggle Kernels

\n\n

Github repos

\n", "rawMarkdown": "This resources list is not exhaustive, it provides just a starting point for Kagglers that would like to join this competition in order to learn, like myself. \n\nI provide here technical articles links, small blog articles, links to Github projects and some inspirational Kaggle Kernels about DeepFake, DCGANs and GANs. \n\n![Wired: DeepFake become better](https://media.wired.com/photos/5ce837fc3cd5de8fe355e337/16:9/w_2278,h_1281,c_limit/Deep-Fake-1905.08233-13.jpg)\n\n**Blogs, Technical Articles**\n\n* Towards Data Science DeepFakes subject selection: [DeepFakes](https://towardsdatascience.com/tagged/deepfakes) \n\n* An introduction to DeepFakes from Towards Data Science: [Deepfakes: The Ugly, and The Good](https://towardsdatascience.com/deepfakes-the-ugly-and-the-good-49115643d8dd) \n\n* Introduction to GAN from Towards Data Science (with code): [GANs from Scratch 1: A deep introduction. With code in PyTorch and TensorFlow](https://medium.com/ai-society/gans-from-scratch-1-a-deep-introduction-with-code-in-pytorch-and-tensorflow-cb03cdcdba0f) \n\n* A cGAN introduction from Mastering Data Science (with code): [How to Develop a Conditional GAN (cGAN) From Scratch](https://machinelearningmastery.com/how-to-develop-a-conditional-generative-adversarial-network-from-scratch/) \n\n* An introduction to GANs on Analytics Vidhya, [GANs — A Brief Introduction to Generative Adversarial Networks](https://medium.com/analytics-vidhya/gans-a-brief-introduction-to-generative-adversarial-networks-f06216c7200e?)\n\n**Tutorials**\n\n* A nice tutorial for GAN using Pytorch: [DCGAN Faces Tutorial](https://pytorch.org/tutorials/beginner/dcgan_faces_tutorial.html) \n\n* A tutorial for GANs: [A Beginner's Guide to Generative Adversarial Networks (GANs)](https://pathmind.com/wiki/generative-adversarial-network-gan) \n\n* Deep Convolutional Generative Adversial Networks, Tensorflow, [DCGAN](https://www.tensorflow.org/tutorials/generative/dcgan) \n\n* A tutorial for video and image colorization and resolution improvement using fast.ai and PyTorch recommended by [@init27](https://www.kaggle.com/init27), [Decrappification, DeOldification, and Super Resolution](https://www.fast.ai/2019/05/03/decrappify/) \n\n* A tutorial from OpenCV for face detection using Cascade Classifiers, [Cascade Classifier](https://docs.opencv.org/3.4/db/d28/tutorial_cascade_classifier.html) \n\n**Kaggle Kernels**\n\n* A very good intro to GAN, by [@nanashi](http://kaggle/nanashi): [GAN Introduction](https://www.kaggle.com/jesucristo/gan-introduction) \n\n* A GAN developed for Dog face generation competition by [@cdeotte](http://kaggle/cdeotte): [Dog Memorizer GAN](https://www.kaggle.com/cdeotte/dog-memorizer-gan) \n\n* A Kernel for Face detection using OpenCV with Haarcascade, by [@serkanpeldek](https://www.kaggle.com/serkanpeldek), [Face Detection with OpenCV](https://www.kaggle.com/serkanpeldek/face-detection-with-opencv) \n\n* Play video and processing, by [@hamditarek](https://www.kaggle.com/hamditarek), https://www.kaggle.com/hamditarek/play-video-and-processing \n\n\n**Github repos**\n\n* Github topic for DeeFakes: [deepfakes](https://github.com/topics/) \n\n* A Github project using Pytorch: [Faceswap-Deepfake-Pytorch](https://github.com/Oldpan/Faceswap-Deepfake-Pytorch) \n\n* A Github project for GAN with PyTorch: [PyTorch-GAN](https://github.com/eriklindernoren/PyTorch-GAN) \n\n* A Github recommended by [@shwetagoyal4](https://www.kaggle.com/shwetagoyal4), [Generative-model-using-PyTorch](https://github.com/Shwetago/Generative-model-using-PyTorch)\n\n* A Github recommended by [@catadanna](https://www.kaggle.com/catadanna) [eriklindernoren/Keras-GAN](https://github.com/eriklindernoren/Keras-GAN)\n", "votes": 95 }, { "id": 1068450, "postDate": "2020-11-03T12:35:04.347Z", "content": "

Just found your topic. I see you mentioned the github repository PyTorch-GAN. I may add, one should follow the author of this repository, the Github user eriklindernoren, who is an ML Engineer at Apple, which means at the very core of GAN research and most probably working with Jan Goodfellow. I prefer this repo too, it contains different GAN architectures.

", "rawMarkdown": "Just found your topic. I see you mentioned the github repository PyTorch-GAN. I may add, one should follow the author of this repository, the Github user `eriklindernoren`, who is an ML Engineer at Apple, which means at the very core of GAN research and most probably working with Jan Goodfellow. I prefer [this repo](https://github.com/eriklindernoren/Keras-GAN) too, it contains different GAN architectures.", "votes": 3 }, { "id": 878862, "postDate": "2020-06-09T00:26:36.417Z", "content": "

Great reference list. Thanks for the post!

", "rawMarkdown": "Great reference list. Thanks for the post! ", "votes": 1 }, { "id": 750816, "postDate": "2020-02-19T17:58:35.443Z", "content": "

Very Helpful!

", "rawMarkdown": "Very Helpful!", "votes": 1 }, { "id": 717641, "postDate": "2020-01-13T12:00:04.483Z", "content": "

@gpreda this is an amazing help for beginners like me! Thank you!

", "rawMarkdown": "@gpreda this is an amazing help for beginners like me! Thank you!\n", "votes": 1 }, { "id": 704270, "postDate": "2019-12-27T08:59:30.867Z", "content": "

Thanks for this.It's very useful for me to get start .

", "rawMarkdown": "Thanks for this.It's very useful for me to get start .", "votes": 1 }, { "id": 703077, "postDate": "2019-12-25T15:01:16.417Z", "content": "

Thanks, It looks good to me, going to start working on it.

", "rawMarkdown": "Thanks, It looks good to me, going to start working on it.", "votes": 1 }, { "id": 696996, "postDate": "2019-12-17T10:44:36.587Z", "content": "

Very useful to me! Thanks for sharing!!😄 👍

", "rawMarkdown": "Very useful to me! Thanks for sharing!!😄 👍 ", "votes": 1 }, { "id": 695216, "postDate": "2019-12-14T19:19:42.780Z", "content": "

Thanks for sharing all these resources!!

\n\n

I have also written a blog and have a GitHub repo, please take a look:\nBlog: GANs - A brief introduction to generative adversarial networks

\n\n

GitHub: Generative model using Pytorch

", "rawMarkdown": "Thanks for sharing all these resources!!\n\nI have also written a blog and have a GitHub repo, please take a look:\nBlog: [GANs - A brief introduction to generative adversarial networks](https://medium.com/analytics-vidhya/gans-a-brief-introduction-to-generative-adversarial-networks-f06216c7200e?)\n\nGitHub: [Generative model using Pytorch](https://github.com/Shwetago/Generative-model-using-PyTorch)", "votes": 1, "replies": [ { "id": 695272, "postDate": "2019-12-14T21:54:32.207Z", "content": "

Thank you for your contribution. I will add to the list.

", "rawMarkdown": "Thank you for your contribution. I will add to the list.", "votes": 1 }, { "id": 695453, "postDate": "2019-12-15T07:08:01.480Z", "content": "

Thank you!! I am glad it helped.

", "rawMarkdown": "Thank you!! I am glad it helped.", "votes": 1 } ] }, { "id": 731674, "postDate": "2020-01-28T22:33:13.900Z", "content": "

Thanks a lot @gpreda and thanks to @nanashi our wunderkind

", "rawMarkdown": "Thanks a lot @gpreda and thanks to @nanashi our wunderkind", "votes": 2, "replies": [ { "id": 731677, "postDate": "2020-01-28T22:38:44.093Z", "content": "

btw, haven't heard much from @hengck23, @cdeotte, @cpmpml, @philippsinger, @iafoss on this challenge. Right up their alley i suppose.

", "rawMarkdown": "btw, haven't heard much from @hengck23, @cdeotte, @cpmpml, @philippsinger, @iafoss on this challenge. Right up their alley i suppose.", "votes": 2 } ] }, { "id": 696122, "postDate": "2019-12-16T06:33:32.693Z", "content": "

@gpreda Thanks for this. It is a great starting point for GAN

", "rawMarkdown": "@gpreda Thanks for this. It is a great starting point for GAN", "votes": 2 }, { "id": 693438, "postDate": "2019-12-12T11:22:00.443Z", "content": "

\"Decrappify\" by fast.ai also has some amazing results.

", "rawMarkdown": "[\"Decrappify\"](https://www.fast.ai/2019/05/03/decrappify/) by fast.ai also has some amazing results. ", "votes": 2, "replies": [ { "id": 693670, "postDate": "2019-12-12T16:56:09.790Z", "content": "

Thank you to contributing to this list, I aim to keep updating as well the list in the following days.

", "rawMarkdown": "Thank you to contributing to this list, I aim to keep updating as well the list in the following days.", "votes": 2 }, { "id": 693999, "postDate": "2019-12-13T02:53:53.373Z", "content": "

Will keep helping you then 😄

", "rawMarkdown": "Will keep helping you then 😄 ", "votes": 2 } ] }, { "id": 700317, "postDate": "2019-12-21T19:31:49.167Z", "content": "

Thanks @gpreda for mentionning my kernel <3

", "rawMarkdown": "Thanks @gpreda for mentionning my kernel <3" }, { "id": 1688321, "postDate": "2022-02-13T15:21:44.183Z", "content": "

Interested fellows can join our Computer Vision research team for research and preparation of interviews regarding #data science #artifical intelligence # image segmentation and classification #multi view learning.

\n

Motivation to share practical and fundamental aspects of implementation and research problems.

\n

CXR,CT,COVID diagnosis,Segmentation and Image classification

\n

Welcome Collaboration
\nHappy Learning to all enthusiastic members as we are sharing knowledge for beginners and advanced level learners.

\n

Join

\n

https://t.me/+XJLDvhp1kQs3NDVl

", "rawMarkdown": "Interested fellows can join our Computer Vision research team for research and preparation of interviews regarding #data science #artifical intelligence # image segmentation and classification #multi view learning.\n\nMotivation to share practical and fundamental aspects of implementation and research problems.\n\n#CXR,CT,COVID diagnosis,Segmentation and Image classification\nWelcome Collaboration\nHappy Learning to all enthusiastic members as we are sharing knowledge for beginners and advanced level learners.\n\n#Join\nhttps://t.me/+XJLDvhp1kQs3NDVl" }, { "id": 1249418, "postDate": "2021-03-23T09:36:29.620Z", "content": "

Very interesting

", "rawMarkdown": "Very interesting" }, { "id": 2836801, "postDate": "2024-05-26T05:57:02.533Z", "rawMarkdown": "", "isDeleted": true }, { "id": 878865, "postDate": "2020-06-09T00:45:59.467Z", "content": "

Good topic, regarding datasets, I found these are pretty relevant:\n- Deepfake-TIMIT\n\n is a faceswap dataset which has 640 videos. The faceswap technique was from the open source GAN-based approach (adapted from here: https://github.com/shaoanlu/faceswap- GAN), which, in turn, was developed from the original autoencoder-based Deepfake al- gorithm (https://github.com/deepfakes/faceswap). When creating the database, we manu- ally selected 16 similar looking pairs of people from publicly available VidTIMIT database (http://conradsanderson.id.au/vidtimit/). For each of 32 subjects, we trained two different mod- els: a lower quality (LQ) with 64 x 64 input/output size model, and higher quality (HQ) with 128 x 128 size model (see the available images for the illustration). Since there are 10 videos per person in VidTIMIT database, we generated 320 videos corresponding to each version, resulting in 620 total videos with faces swapped. For the audio, we kept the original audio track of each video, i.e., no manipulation was done to the audio channel.\n\n- Celeb-DF-v2\n- Face-Forensics++

\n\n

my download choice\npython faceforensics_download_v4.py /Users/dph/downloads/data-ff -d all -c c40 -t videos -n 200

\n\n\n\n

But what is the benchmark for anyone want to keep researching deepfake?

", "rawMarkdown": "Good topic, regarding datasets, I found these are pretty relevant:\n- Deepfake-TIMIT\n```\n is a faceswap dataset which has 640 videos. The faceswap technique was from the open source GAN-based approach (adapted from here: https://github.com/shaoanlu/faceswap- GAN), which, in turn, was developed from the original autoencoder-based Deepfake al- gorithm (https://github.com/deepfakes/faceswap). When creating the database, we manu- ally selected 16 similar looking pairs of people from publicly available VidTIMIT database (http://conradsanderson.id.au/vidtimit/). For each of 32 subjects, we trained two different mod- els: a lower quality (LQ) with 64 x 64 input/output size model, and higher quality (HQ) with 128 x 128 size model (see the available images for the illustration). Since there are 10 videos per person in VidTIMIT database, we generated 320 videos corresponding to each version, resulting in 620 total videos with faces swapped. For the audio, we kept the original audio track of each video, i.e., no manipulation was done to the audio channel.\n```\n- Celeb-DF-v2\n- [Face-Forensics++](https://github.com/ondyari/FaceForensics/blob/master/dataset/README.md )\n\nmy download choice\n`python faceforensics_download_v4.py /Users/dph/downloads/data-ff -d all -c c40 -t videos -n 200`\n\n- [ Kaggle-real-fake-faces](https://www.kaggle.com/ciplab/real-and-fake-face-detection)\n- [DFDC](https://www.kaggle.com/c/deepfake-detection-challenge/data)\n\n\n\nBut what is the benchmark for anyone want to keep researching deepfake?", "isDeleted": true }, { "id": 751308, "postDate": "2020-02-20T05:07:19.613Z", "content": "

Interesting links! Thanks for sharing.

", "rawMarkdown": "Interesting links! Thanks for sharing.", "votes": 1 }, { "id": 703100, "postDate": "2019-12-25T15:44:51.583Z", "content": "

Thanks for share! 💯

", "rawMarkdown": "Thanks for share! 💯 ", "votes": 1 }, { "id": 695275, "postDate": "2019-12-14T22:14:08.907Z", "content": "

Thanks for the post

", "rawMarkdown": "Thanks for the post", "votes": 1 }, { "id": 695053, "postDate": "2019-12-14T13:40:32.953Z", "content": "

Thank you for the post!

", "rawMarkdown": "Thank you for the post!", "votes": 1 }, { "id": 717768, "postDate": "2020-01-13T15:44:03.877Z", "content": "

Thanks for this resource

", "rawMarkdown": "Thanks for this resource", "votes": 2 }, { "id": 695611, "postDate": "2019-12-15T11:42:40.237Z", "content": "

@gpreda thanks for sharing this..

", "rawMarkdown": "@gpreda thanks for sharing this..", "votes": 2 }, { "id": 1494664, "postDate": "2021-08-28T23:04:50.437Z", "content": "

Thanks for such an exhaustive list!

", "rawMarkdown": "Thanks for such an exhaustive list!" }, { "id": 754083, "postDate": "2020-02-23T03:10:59.587Z", "content": "

Thank you for sharing

", "rawMarkdown": "Thank you for sharing", "votes": 1, "isDeleted": true } ], "comments": [ { "id": 1068450, "author_name": "Catadanna", "author_url": "", "post_date": "2020-11-03T12:35:04.347000", "content": "

Just found your topic. I see you mentioned the github repository PyTorch-GAN. I may add, one should follow the author of this repository, the Github user eriklindernoren, who is an ML Engineer at Apple, which means at the very core of GAN research and most probably working with Jan Goodfellow. I prefer this repo too, it contains different GAN architectures.

", "votes": 3, "replies": [] }, { "id": 878862, "author_name": "Seth Hamilton", "author_url": "", "post_date": "2020-06-09T00:26:36.417000", "content": "

Great reference list. Thanks for the post!

", "votes": 1, "replies": [] }, { "id": 750816, "author_name": "Akash", "author_url": "", "post_date": "2020-02-19T17:58:35.443000", "content": "

Very Helpful!

", "votes": 1, "replies": [] }, { "id": 717641, "author_name": "Raluca Saru", "author_url": "", "post_date": "2020-01-13T12:00:04.483000", "content": "

@gpreda this is an amazing help for beginners like me! Thank you!

", "votes": 1, "replies": [] }, { "id": 704270, "author_name": "", "author_url": "", "post_date": "2019-12-27T08:59:30.867000", "content": "

Thanks for this.It's very useful for me to get start .

", "votes": 1, "replies": [] }, { "id": 703077, "author_name": "Manish Kumar Singh", "author_url": "", "post_date": "2019-12-25T15:01:16.417000", "content": "

Thanks, It looks good to me, going to start working on it.

", "votes": 1, "replies": [] }, { "id": 696996, "author_name": "Miyabon", "author_url": "", "post_date": "2019-12-17T10:44:36.587000", "content": "

Very useful to me! Thanks for sharing!!😄 👍

", "votes": 1, "replies": [] }, { "id": 695216, "author_name": "Shweta Goyal", "author_url": "", "post_date": "2019-12-14T19:19:42.780000", "content": "

Thanks for sharing all these resources!!

\n\n

I have also written a blog and have a GitHub repo, please take a look:\nBlog: GANs - A brief introduction to generative adversarial networks

\n\n

GitHub: Generative model using Pytorch

", "votes": 1, "replies": [ { "id": 695272, "author_name": "Gabriel Preda", "author_url": "", "post_date": "2019-12-14T21:54:32.207000", "content": "

Thank you for your contribution. I will add to the list.

", "votes": 1, "replies": [] }, { "id": 695453, "author_name": "Shweta Goyal", "author_url": "", "post_date": "2019-12-15T07:08:01.480000", "content": "

Thank you!! I am glad it helped.

", "votes": 1, "replies": [] } ] }, { "id": 731674, "author_name": "student", "author_url": "", "post_date": "2020-01-28T22:33:13.900000", "content": "

Thanks a lot @gpreda and thanks to @nanashi our wunderkind

", "votes": 2, "replies": [ { "id": 731677, "author_name": "student", "author_url": "", "post_date": "2020-01-28T22:38:44.093000", "content": "

btw, haven't heard much from @hengck23, @cdeotte, @cpmpml, @philippsinger, @iafoss on this challenge. Right up their alley i suppose.

", "votes": 2, "replies": [] } ] }, { "id": 696122, "author_name": "Sandeep_pandu", "author_url": "", "post_date": "2019-12-16T06:33:32.693000", "content": "

@gpreda Thanks for this. It is a great starting point for GAN

", "votes": 2, "replies": [] }, { "id": 693438, "author_name": "Sanyam Bhutani", "author_url": "", "post_date": "2019-12-12T11:22:00.443000", "content": "

\"Decrappify\" by fast.ai also has some amazing results.

", "votes": 2, "replies": [ { "id": 693670, "author_name": "Gabriel Preda", "author_url": "", "post_date": "2019-12-12T16:56:09.790000", "content": "

Thank you to contributing to this list, I aim to keep updating as well the list in the following days.

", "votes": 2, "replies": [] }, { "id": 693999, "author_name": "Sanyam Bhutani", "author_url": "", "post_date": "2019-12-13T02:53:53.373000", "content": "

Will keep helping you then 😄

", "votes": 2, "replies": [] } ] }, { "id": 700317, "author_name": "Tarek Hamdi", "author_url": "", "post_date": "2019-12-21T19:31:49.167000", "content": "

Thanks @gpreda for mentionning my kernel <3

", "votes": 0, "replies": [] }, { "id": 1688321, "author_name": "Ajay sharma", "author_url": "", "post_date": "2022-02-13T15:21:44.183000", "content": "

Interested fellows can join our Computer Vision research team for research and preparation of interviews regarding #data science #artifical intelligence # image segmentation and classification #multi view learning.

\n

Motivation to share practical and fundamental aspects of implementation and research problems.

\n

CXR,CT,COVID diagnosis,Segmentation and Image classification

\n

Welcome Collaboration
\nHappy Learning to all enthusiastic members as we are sharing knowledge for beginners and advanced level learners.

\n

Join

\n

https://t.me/+XJLDvhp1kQs3NDVl

", "votes": 0, "replies": [] }, { "id": 1249418, "author_name": "Amir Bouzidi", "author_url": "", "post_date": "2021-03-23T09:36:29.620000", "content": "

Very interesting

", "votes": 0, "replies": [] }, { "id": 2836801, "author_name": "", "author_url": "", "post_date": "2024-05-26T05:57:02.533000", "content": "", "votes": 0, "replies": [] }, { "id": 878865, "author_name": "", "author_url": "", "post_date": "2020-06-09T00:45:59.467000", "content": "

Good topic, regarding datasets, I found these are pretty relevant:\n- Deepfake-TIMIT\n\n is a faceswap dataset which has 640 videos. The faceswap technique was from the open source GAN-based approach (adapted from here: https://github.com/shaoanlu/faceswap- GAN), which, in turn, was developed from the original autoencoder-based Deepfake al- gorithm (https://github.com/deepfakes/faceswap). When creating the database, we manu- ally selected 16 similar looking pairs of people from publicly available VidTIMIT database (http://conradsanderson.id.au/vidtimit/). For each of 32 subjects, we trained two different mod- els: a lower quality (LQ) with 64 x 64 input/output size model, and higher quality (HQ) with 128 x 128 size model (see the available images for the illustration). Since there are 10 videos per person in VidTIMIT database, we generated 320 videos corresponding to each version, resulting in 620 total videos with faces swapped. For the audio, we kept the original audio track of each video, i.e., no manipulation was done to the audio channel.\n\n- Celeb-DF-v2\n- Face-Forensics++

\n\n

my download choice\npython faceforensics_download_v4.py /Users/dph/downloads/data-ff -d all -c c40 -t videos -n 200

\n\n\n\n

But what is the benchmark for anyone want to keep researching deepfake?

", "votes": 0, "replies": [] }, { "id": 751308, "author_name": "SkyLord", "author_url": "", "post_date": "2020-02-20T05:07:19.613000", "content": "

Interesting links! Thanks for sharing.

", "votes": 1, "replies": [] }, { "id": 703100, "author_name": "dasmehdixtr", "author_url": "", "post_date": "2019-12-25T15:44:51.583000", "content": "

Thanks for share! 💯

", "votes": 1, "replies": [] }, { "id": 695275, "author_name": "AtulVerma", "author_url": "", "post_date": "2019-12-14T22:14:08.907000", "content": "

Thanks for the post

", "votes": 1, "replies": [] }, { "id": 695053, "author_name": "Beans", "author_url": "", "post_date": "2019-12-14T13:40:32.953000", "content": "

Thank you for the post!

", "votes": 1, "replies": [] }, { "id": 717768, "author_name": "Sai Srinivas Reddy", "author_url": "", "post_date": "2020-01-13T15:44:03.877000", "content": "

Thanks for this resource

", "votes": 2, "replies": [] }, { "id": 695611, "author_name": "Navneet Kumar", "author_url": "", "post_date": "2019-12-15T11:42:40.237000", "content": "

@gpreda thanks for sharing this..

", "votes": 2, "replies": [] }, { "id": 1494664, "author_name": "Tanuj Sharma", "author_url": "", "post_date": "2021-08-28T23:04:50.437000", "content": "

Thanks for such an exhaustive list!

", "votes": 0, "replies": [] }, { "id": 754083, "author_name": "", "author_url": "", "post_date": "2020-02-23T03:10:59.587000", "content": "

Thank you for sharing

", "votes": 1, "replies": [] } ], "raw_markdown_by_id": { "693424": "This resources list is not exhaustive, it provides just a starting point for Kagglers that would like to join this competition in order to learn, like myself. \n\nI provide here technical articles links, small blog articles, links to Github projects and some inspirational Kaggle Kernels about DeepFake, DCGANs and GANs. \n\n![Wired: DeepFake become better](https://media.wired.com/photos/5ce837fc3cd5de8fe355e337/16:9/w_2278,h_1281,c_limit/Deep-Fake-1905.08233-13.jpg)\n\n**Blogs, Technical Articles**\n\n* Towards Data Science DeepFakes subject selection: [DeepFakes](https://towardsdatascience.com/tagged/deepfakes) \n\n* An introduction to DeepFakes from Towards Data Science: [Deepfakes: The Ugly, and The Good](https://towardsdatascience.com/deepfakes-the-ugly-and-the-good-49115643d8dd) \n\n* Introduction to GAN from Towards Data Science (with code): [GANs from Scratch 1: A deep introduction. With code in PyTorch and TensorFlow](https://medium.com/ai-society/gans-from-scratch-1-a-deep-introduction-with-code-in-pytorch-and-tensorflow-cb03cdcdba0f) \n\n* A cGAN introduction from Mastering Data Science (with code): [How to Develop a Conditional GAN (cGAN) From Scratch](https://machinelearningmastery.com/how-to-develop-a-conditional-generative-adversarial-network-from-scratch/) \n\n* An introduction to GANs on Analytics Vidhya, [GANs — A Brief Introduction to Generative Adversarial Networks](https://medium.com/analytics-vidhya/gans-a-brief-introduction-to-generative-adversarial-networks-f06216c7200e?)\n\n**Tutorials**\n\n* A nice tutorial for GAN using Pytorch: [DCGAN Faces Tutorial](https://pytorch.org/tutorials/beginner/dcgan_faces_tutorial.html) \n\n* A tutorial for GANs: [A Beginner's Guide to Generative Adversarial Networks (GANs)](https://pathmind.com/wiki/generative-adversarial-network-gan) \n\n* Deep Convolutional Generative Adversial Networks, Tensorflow, [DCGAN](https://www.tensorflow.org/tutorials/generative/dcgan) \n\n* A tutorial for video and image colorization and resolution improvement using fast.ai and PyTorch recommended by [@init27](https://www.kaggle.com/init27), [Decrappification, DeOldification, and Super Resolution](https://www.fast.ai/2019/05/03/decrappify/) \n\n* A tutorial from OpenCV for face detection using Cascade Classifiers, [Cascade Classifier](https://docs.opencv.org/3.4/db/d28/tutorial_cascade_classifier.html) \n\n**Kaggle Kernels**\n\n* A very good intro to GAN, by [@nanashi](http://kaggle/nanashi): [GAN Introduction](https://www.kaggle.com/jesucristo/gan-introduction) \n\n* A GAN developed for Dog face generation competition by [@cdeotte](http://kaggle/cdeotte): [Dog Memorizer GAN](https://www.kaggle.com/cdeotte/dog-memorizer-gan) \n\n* A Kernel for Face detection using OpenCV with Haarcascade, by [@serkanpeldek](https://www.kaggle.com/serkanpeldek), [Face Detection with OpenCV](https://www.kaggle.com/serkanpeldek/face-detection-with-opencv) \n\n* Play video and processing, by [@hamditarek](https://www.kaggle.com/hamditarek), https://www.kaggle.com/hamditarek/play-video-and-processing \n\n\n**Github repos**\n\n* Github topic for DeeFakes: [deepfakes](https://github.com/topics/) \n\n* A Github project using Pytorch: [Faceswap-Deepfake-Pytorch](https://github.com/Oldpan/Faceswap-Deepfake-Pytorch) \n\n* A Github project for GAN with PyTorch: [PyTorch-GAN](https://github.com/eriklindernoren/PyTorch-GAN) \n\n* A Github recommended by [@shwetagoyal4](https://www.kaggle.com/shwetagoyal4), [Generative-model-using-PyTorch](https://github.com/Shwetago/Generative-model-using-PyTorch)\n\n* A Github recommended by [@catadanna](https://www.kaggle.com/catadanna) [eriklindernoren/Keras-GAN](https://github.com/eriklindernoren/Keras-GAN)\n", "1068450": "Just found your topic. I see you mentioned the github repository PyTorch-GAN. I may add, one should follow the author of this repository, the Github user `eriklindernoren`, who is an ML Engineer at Apple, which means at the very core of GAN research and most probably working with Jan Goodfellow. I prefer [this repo](https://github.com/eriklindernoren/Keras-GAN) too, it contains different GAN architectures.", "878862": "Great reference list. Thanks for the post! ", "750816": "Very Helpful!", "717641": "@gpreda this is an amazing help for beginners like me! Thank you!\n", "704270": "Thanks for this.It's very useful for me to get start .", "703077": "Thanks, It looks good to me, going to start working on it.", "696996": "Very useful to me! Thanks for sharing!!😄 👍 ", "695216": "Thanks for sharing all these resources!!\n\nI have also written a blog and have a GitHub repo, please take a look:\nBlog: [GANs - A brief introduction to generative adversarial networks](https://medium.com/analytics-vidhya/gans-a-brief-introduction-to-generative-adversarial-networks-f06216c7200e?)\n\nGitHub: [Generative model using Pytorch](https://github.com/Shwetago/Generative-model-using-PyTorch)", "731674": "Thanks a lot @gpreda and thanks to @nanashi our wunderkind", "696122": "@gpreda Thanks for this. It is a great starting point for GAN", "693438": "[\"Decrappify\"](https://www.fast.ai/2019/05/03/decrappify/) by fast.ai also has some amazing results. ", "700317": "Thanks @gpreda for mentionning my kernel <3", "1688321": "Interested fellows can join our Computer Vision research team for research and preparation of interviews regarding #data science #artifical intelligence # image segmentation and classification #multi view learning.\n\nMotivation to share practical and fundamental aspects of implementation and research problems.\n\n#CXR,CT,COVID diagnosis,Segmentation and Image classification\nWelcome Collaboration\nHappy Learning to all enthusiastic members as we are sharing knowledge for beginners and advanced level learners.\n\n#Join\nhttps://t.me/+XJLDvhp1kQs3NDVl", "1249418": "Very interesting", "2836801": "", "878865": "Good topic, regarding datasets, I found these are pretty relevant:\n- Deepfake-TIMIT\n```\n is a faceswap dataset which has 640 videos. The faceswap technique was from the open source GAN-based approach (adapted from here: https://github.com/shaoanlu/faceswap- GAN), which, in turn, was developed from the original autoencoder-based Deepfake al- gorithm (https://github.com/deepfakes/faceswap). When creating the database, we manu- ally selected 16 similar looking pairs of people from publicly available VidTIMIT database (http://conradsanderson.id.au/vidtimit/). For each of 32 subjects, we trained two different mod- els: a lower quality (LQ) with 64 x 64 input/output size model, and higher quality (HQ) with 128 x 128 size model (see the available images for the illustration). Since there are 10 videos per person in VidTIMIT database, we generated 320 videos corresponding to each version, resulting in 620 total videos with faces swapped. For the audio, we kept the original audio track of each video, i.e., no manipulation was done to the audio channel.\n```\n- Celeb-DF-v2\n- [Face-Forensics++](https://github.com/ondyari/FaceForensics/blob/master/dataset/README.md )\n\nmy download choice\n`python faceforensics_download_v4.py /Users/dph/downloads/data-ff -d all -c c40 -t videos -n 200`\n\n- [ Kaggle-real-fake-faces](https://www.kaggle.com/ciplab/real-and-fake-face-detection)\n- [DFDC](https://www.kaggle.com/c/deepfake-detection-challenge/data)\n\n\n\nBut what is the benchmark for anyone want to keep researching deepfake?", "751308": "Interesting links! Thanks for sharing.", "703100": "Thanks for share! 💯 ", "695275": "Thanks for the post", "695053": "Thank you for the post!", "717768": "Thanks for this resource", "695611": "@gpreda thanks for sharing this..", "1494664": "Thanks for such an exhaustive list!", "754083": "Thank you for sharing" } }