98 - Open Talk - How to detect deepfake

37:17
 
シェア
 

Manage episode 301652425 series 2922369
著作 Kashif Manzoor の情報はPlayer FM及びコミュニティによって発見されました。著作権は出版社によって所持されます。そして、番組のオーディオは、その出版社のサーバから直接にストリーミングされます。Player FMで購読ボタンをタップし、更新できて、または他のポッドキャストアプリにフィードのURLを貼り付けます。

A deepfake refers to a specific kind of media or photo where a person in an image or video is transacted with another person however it will look like the original person in an image or video.

To make a deep fake video, a creator swaps one person’s face and replaces it with another, using a deep learning algorithm generative adversarial networks (GAN).

A formal definition as per Wikipedia:

Deepfakes are synthetic media in which a person in an existing image or video is replaced with someone else's likeness. While the act of faking content is not new, deepfakes leverage powerful techniques from machine learning and artificial intelligence to manipulate or generate visual and audio content with a high potential to deceive. The main machine learning methods used to create deepfakes are based on deep learning and involve training generative neural network architectures, such as autoencoders or generative adversarial networks (GANs) – Source: Wikipedia

Today’s Guest

Ilke Demir, Research Scientist at Intel

An inspirational talk with the guest speaker on a variety of topics, like

  • How to decide to pursue a short course or a degree program in AI/Data Science,
  • What is a new fake catcher tool? and from where you can find out the number of large datasets.

Resources:

105 つのエピソード