DeepDream

Mona Liza sa DeepDream efektom koristeći VGG16 mrežu obučenu na ImageNet-u

DeepDream je program za računarski vid kreiran od strane Guglovih inženjera Aleksandra Mordvinceva koji koristi konvolucionu neuronsku mrežu da pronađe i poboljša paterne na slikama putem algoritamske pareidolije, stvarajući tako izgled nalik snu koji podseća na psihodelično iskustvo u namerno prerađenim slikama.[1][2][3]

Guglov program je popularizovao termin (duboko) „sanjanje“ da se odnosi na generisanje slika koje proizvode željene aktivacije u obučenoj dubokoj mreži, a termin se sada odnosi na kolekciju srodnih pristupa.

Istorija

Softver DeepDream, nastao u dubokoj konvolucionoj mreži pod kodnim imenom „Inception“ po istoimenom filmu,[1][2][3] je razvijen za ImageNet veliki izazov vizuelnog prepoznavanja (ILSVRC) 2014. godine[3] i objavljen u julu 2015.

Ideja i ime iz snova postali su popularni na internetu 2015. zahvaljujući Guglovom programu Dipdrim. Ideja datira iz rane istorije neuronskih mreža,[4] i slične metode su korišćene za sintezu vizuelnih tekstura.[5] Srodne ideje za vizuelizaciju razvilo je (pre Guglovog rada) nekoliko istraživačkih grupa.[6][7]

Nakon što je Gugl objavio svoje tehnike i učinio svoj kod otvorenim kodom,[8] na tržištu su se pojavili brojni alati u obliku veb servisa, mobilnih aplikacija i desktop softvera koji omogućavaju korisnicima da transformišu sopstvene fotografije.[9]

Reference

  1. ^ а б Mordvintsev, Alexander; Olah, Christopher; Tyka, Mike (2015). „DeepDream - a code example for visualizing Neural Networks”. Google Research. Архивирано из оригинала 2015-07-08. г.. 
  2. ^ а б Mordvintsev, Alexander; Olah, Christopher; Tyka, Mike (2015). „Inceptionism: Going Deeper into Neural Networks”. Google Research. Архивирано из оригинала 2015-07-03. г.. 
  3. ^ а б в Szegedy, Christian; Liu, Wei; Jia, Yangqing; Sermanet, Pierre; Reed, Scott E.; Anguelov, Dragomir; Erhan, Dumitru; Vanhoucke, Vincent; Rabinovich, Andrew (2015). „Going deeper with convolutions”. IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015, Boston, MA, USA, June 7–12, 2015. IEEE Computer Society. стр. 1—9. arXiv:1409.4842 Слободан приступ. doi:10.1109/CVPR.2015.7298594. 
  4. ^ Lewis, J.P. (1988). „Creation by refinement: a creativity paradigm for gradient descent learning networks”. IEEE International Conference on Neural Networks. IEEE International Conference on Neural Networks. стр. 229—233 vol.2. ISBN 0-7803-0999-5. doi:10.1109/ICNN.1988.23933. 
  5. ^ Portilla, J; Simoncelli, Eero (2000). „A parametric texture model based on joint statistics of complex wavelet coefficients”. International Journal of Computer Vision. 40: 49—70. S2CID 2475577. doi:10.1023/A:1026553619983. 
  6. ^ Erhan, Dumitru. (2009). Visualizing Higher-Layer Features of a Deep Network. International Conference on Machine Learning Workshop on Learning Feature Hierarchies. S2CID 15127402. 
  7. ^ Simonyan, Karen; Vedaldi, Andrea; Zisserman, Andrew (2014). Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps. International Conference on Learning Representations Workshop. arXiv:1312.6034 Слободан приступ. 
  8. ^ deepdream на веб-сајту GitHub
  9. ^ Daniel Culpan (2015-07-03). „These Google "Deep Dream" Images Are Weirdly Mesmerising”. Wired. Архивирано из оригинала 08. 05. 2016. г. Приступљено 2015-07-25. 

Spoljašnje veze

DeepDream на Викимедијиној остави.
Spoljašnji video-zapis
Deep Dream (Google) - Computerphile by Michael Pound

Content Disclaimer

Informasi ini disarikan dari Wikipedia dan disajikan kembali untuk tujuan edukasi. Konten tersedia di bawah lisensi CC BY-SA 3.0. Kami tidak bertanggung jawab atas ketidakakuratan data yang bersumber dari kontribusi publik tersebut.

  1. The information displayed on this website is sourced in part or in whole from Wikipedia and has been adapted for the purpose of restating it. We strive to provide accurate and relevant information, however:
  2. There is no guarantee of absolute accuracy. Wikipedia is an open, collaborative project that can be edited by anyone, so information is subject to change.
  3. It is not intended to constitute professional advice. The content displayed is for informational and educational purposes only. For important decisions (e.g., medical, legal, or financial), please consult a professional.
  4. Content copyright. Wikipedia is licensed under the Creative Commons Attribution-ShareAlike License (CC BY-SA). This means that content may be reused with appropriate attribution and shared under a similar license.
  5. Responsible use. Any risk arising from the use of information from this website is entirely the responsibility of the user.