Recurrent Neural Network Writes Sentences About Images | Two Minute Papers #23 Video
This technique is a combination of two powerful machine learning algorithms:
- convolutional neural networks are excellent at image classification, i.e., finding out what is seen on an input image,
- recurrent neural networks that are capable of processing a sequence of inputs and outputs, therefore it can create sentences of what is seen on the image.
Combining these two techniques makes it possible for a computer to describe in a sentence what is seen on an input image.
_____________________
The paper "Deep Visual-Semantic Alignments for Generating Image Descriptions" is available here:
http://cs.stanford.edu/people/karpathy/deepimagesent/
A gallery with more results with the same algorithm:
http://cs.stanford.edu/people/karpathy/deepimagesent/generationdemo/
You can train your own convolutional neural network here:
http://cs.stanford.edu/people/karpathy/convnetjs/demo/cifar10.html
The source code for the project is now available here:
https://github.com/karpathy/neuraltalk2
Subscribe if you would like to see more of these! - http://www.youtube.com/subscription_center?add_user=keeroyz
The thumbnail image background was made by Georgie Pauwels (CC BY 2.0) - https://flic.kr/p/qrRciQ
Splash screen/thumbnail design: Felícia Fehér - http://felicia.hu
Károly Zsolnai-Fehér's links:
Patreon → https://www.patreon.com/TwoMinutePapers
Facebook → https://www.facebook.com/TwoMinutePapers/
Twitter → https://twitter.com/karoly_zsolnai
Web → https://cg.tuwien.ac.at/~zsolnai/
About the Site 🌐
This site provides links to random videos hosted at YouTube, with the emphasis on random. 🎥
Origins of the Idea 🌱
The original idea for this site stemmed from the need to benchmark the popularity of a video against the general population of YouTube videos. 🧠
Challenges Faced 🤔
Obtaining a large sample of videos was crucial for accurate ranking, but YouTube lacks a direct method to gather random video IDs.
Even searching for random strings on YouTube doesn't yield truly random results, complicating the process further. 🔍
Creating Truly Random Links 🛠️
The YouTube API offers additional functions enabling the discovery of more random videos. Through inventive techniques and a touch of space-time manipulation, we've achieved a process yielding nearly 100% random links to YouTube videos.
About YouTube 📺
YouTube, an American video-sharing website based in San Bruno, California, offers a diverse range of user-generated and corporate media content. 🌟
Content and Users 🎵
Users can upload, view, rate, share, and comment on videos, with content spanning video clips, music videos, live streams, and more.
While most content is uploaded by individuals, media corporations like CBS and the BBC also contribute. Unregistered users can watch videos, while registered users enjoy additional privileges such as uploading unlimited videos and adding comments.
Monetization and Impact 🤑
YouTube and creators earn revenue through Google AdSense, with most videos free to view. Premium channels and subscription services like YouTube Music and YouTube Premium offer ad-free streaming.
As of February 2017, over 400 hours of content were uploaded to YouTube every minute, with the site ranking as the second-most popular globally. By May 2019, this figure exceeded 500 hours per minute. 📈
List of ours generators⚡
Random YouTube Videos Generator
Random Film and Animation Video Generator
Random Autos and Vehicles Video Generator
Random Pets and Animals Video Generator
Random Travel and Events Video Generator
Random People and Blogs Video Generator
Random Entertainment Video Generator
Random News and Politics Video Generator
Random Howto and Style Video Generator
Random Education Video Generator