Verena: Unterschied zwischen den Versionen
Aus exmediawiki
Verena (Diskussion | Beiträge) |
Verena (Diskussion | Beiträge) |
||
Zeile 29: | Zeile 29: | ||
===Technische Umsetzung=== | ===Technische Umsetzung=== | ||
+ | |||
+ | ====Erste Schritte==== | ||
+ | |||
+ | - Convolutional Neural Networks | ||
+ | - Heartbeat Tutorial: [https://heartbeat.fritz.ai/analyze-and-visualize-detected-video-objects-using-keras-and-imageai-d84c99b0ae8e Analyze and Visualize Detected Video Objects Using Keras and ImageAI] | ||
====Aktueller Stand==== | ====Aktueller Stand==== |
Version vom 17. Mai 2019, 17:18 Uhr
...
Inhaltsverzeichnis
KI-SEMINAR
Projektdokumentation
Kurzbeschreibung (EN)
THE OFFICE (working title), 2019
Humans and plants live on different timescales. This is certainly one of the reasons
why in everyday life, plants might often seem static and object-like to us.
»The Office« (2019) makes use of existing video footage that covers large time
spans, in this case, popular long-term tv shows. These productions showcase office
and apartment interiors over a longer period of time, often also including houseplants
which are mainly used for decorative purposes. Some of these series run over
years or even decades and thus comprise many hours of footage.
Convolutional neural networks are used as a tool to detect scenes involving houseplants
while processing large quantities of the given video material. Selected scenes
are compiled into a time-lapse movie which documents plant growth over a long
period of time. While the lively movement and growth of the normally passively
seeming plants becomes visible, the human activities become blurry and fade into
the background.
Keywords: plants, plant-human-relationships, time, timescales, time-lapse, video,
neural networks
Hintergrund/Research
Technische Umsetzung
Erste Schritte
- Convolutional Neural Networks - Heartbeat Tutorial: Analyze and Visualize Detected Video Objects Using Keras and ImageAI
Aktueller Stand
Nächste Schritte
Links zum Thema
Evolutionary Algorithms
- Evolutionary algorithm outperforms deep-learning machines at video games, MIT Technology Review, 06/2018
Code
BLOCKCHAIN READING GROUP notes
> seminar page: Blockchain Reading Group
Bitcoin & Blockchain Basics
- What is Bitcoin? A Beginner’s Guide, coinsutra, 10/2018
- Blockchain Explorer (live)
Ethereum
- We Met The Founder Of Ethereum, VICE on HBO, 04/2018
- "Code For Ethereum’s Proof-of-Stake Blockchain to Be Finalized Next Month", Coindesk, 05/2019
- currently transitioning from proof-of-work to proof-of-stake (PoS) blockchain
- Vitalik Buterin, recent interview "Decentralized finance is going to come first", youtube, 04/2019
- smart contract based insurance, e.g. if a flood happens you automatically get a payout, HurricaneGuard, crop insurance (Sri Lanka)
- social recovery approach (e.g. 5 people sharing one key)
- Ethereum Cryptocurrency: Everything A Beginner Needs To Know, coinsutra, 09/2018
- "Ethereum is not just a blockchain; it’s a decentralized programmable blockchain-based software platform."
- https://en.wikipedia.org/wiki/Smart_contract
- 6 Interesting Blockchain Projects
KünstlerInnenhonorare
- Linksammlung zum Thema KünstlerInnenhonorare
Other Research
Teaching & Non-Teaching
- Meet the school with no classes, no classrooms and no curriculum, Medium, 05/2019