"I am AI" Docuseries, Episode 9: A Bountiful Harvest - Agrobot: #ML #Regenerative #FFHCI
"I am AI" Docuseries, Episode 9: A Bountiful Harvest - Agrobot: #ML #Regenerative #FFHCI
SFV: Reinforcement Learning of Physical Skills from Videos: #ML
https://arxiv.org/abs/1810.03599v1
"MIDI-VAE: Modeling Dynamics and Instrumentation of Music with Applications to Style Transfer": https://arxiv.org/abs/1809.07600 code: https://github.com/brunnergino/MIDI-VAE
"Symbolic Music Genre Transfer with CycleGAN": https://arxiv.org/abs/1809.07575 code: https://github.com/sumuzhao/CycleGAN-Music-Style-Transfer #Generative #Music #ML
"Attribute Enhanced Face Aging with Wavelet-based Generative Adversarial Networks":
https://arxiv.org/abs/1809.06647v1 #ML #Generative
Neural Animation and Reenactment of Human Actor Videos:
http://gvv.mpi-inf.mpg.de/projects/wxu/HumanReenactment/
Human-AI Collaborated Graffiti:
https://howtogeneratealmostanything.com/graffiti/2018/09/26/episode5.html
Youtube-8M: https://research.google.com/youtube8m/ #ML
https://research.google.com/youtube8m/explore.html
YouTube-8M is a large-scale labeled video dataset that consists of millions of YouTube video IDs, with high-quality machine-generated annotations from a diverse vocabulary of 3,800+ visual entities. It comes with precomputed audio-visual features from billions of frames and audio segments, designed to fit on a single hard disk. This makes it possible to train a strong baseline model on this dataset in less than a day on a single GPU! At the same time, the dataset's scale and diversity can enable deep exploration of complex audio-visual models that can take weeks to train even in a distributed fashion.
"Visualizing and Understanding Generative Adversarial Networks":
BrainNet: A Multi-Person Brain-to-Brain Interface for Direct Collaboration Between Brains:
https://arxiv.org/abs/1809.08632v1 #BCI #NeuroScience #ML
Semantic Soft Segmentation (SIGGRAPH 2018)
http://cfg.mit.edu/sites/cfg.mit.edu/files/sss_3.pdf
A Future with Generative Design - talk by Morgan Fabian (Autodesk)
Generative Design - talk by Erin Bradner (Autodesk)
Generative Design: Co-creating with A.I - talk by Erin Bradner (Autodesk)
Visual Memory QA - interesting work on semantic video search. #ML
Presentation on "Grammar Variational Autoencoder"
"Visual Rhythm and Beat" - SIGGRAPH 2018 work by @AbeDavis on manipulating dance in Video: http://abedavis.com/visualbeat/ #Generative #ML
Recorded Future - Intelligence-Driven Security: https://www.recordedfuture.com/.
https://en.wikipedia.org/wiki/Recorded_Future "founded in 2009, specializing in real-time threat intelligence. Has close links with In-Q-Tel, CIA’s investment arm, and Google Ventures."
Talk by Recorded Future CEO Christopher Ahlberg:
The reason why people are fine with "templates" that are highly constrained in their possible creative output styles (& don't need truly creative generative "AI's"), is the fact that most creative fields adhere to fashion cycles: You want something similar to X, not something new.
The majority of the commercial "AI for creative field X" offerings have essentially figured out how to do clever use of dynamic templates, combined with massive media libraries. The results are often impressive but have very little todo with "AI", beyond a useful marketing term.
"Knowledge Graphs & Deep Learning at YouTube". #ML
Knowledge Systems and AI. #ML