innovator

Add idea


Calendar

«    September 2019    »
MonTueWedThuFriSatSun
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
 

 

Advert

 

Payment

 

Advert

 

Authorization

Стартап

DateDate: 11-09-2019, 05:52



Personal non-property and property rights for the invention belong to the author and protected under current legislation.
 The idea is that the base of the hookah mounted compact pump, which allows using both automatic and remote control to produce it run. That is, when smoking a hookah a person is going to inhale the smoke, he presses a button on the remote or the button on the base tube. Runs, respectively, the pump, and the smoke from shisha is automatically available to the smoker.  Technically possible, if necessary, you can install a floating adjustment of supply power to the pump that will allow you to change the feed speed of fragrant smoke the smoker. 
 This idea allows to simplify the procedure of hookah smoking. Also, in the future, it is possible to run mobile apps. 
 The idea is open for investment and complete its foreclosure.
 +380505238948

DateDate: 11-09-2019, 05:48

The neural network was taught to transform the image into music.
Researchers from the University of Amsterdam presented a neural network that can relate visual sensations to sounds - and turn pictures into music. The work was published in an article on arXiv.org, and will be presented at the ICCVW 2019 conference.
When teaching the algorithm, the researchers did not show him how to correlate the image with the music - the neural network received these skills on its own using the teaching method without a teacher.
The algorithm is built on the principle of an auto-encoder and consists of an encoder and a decoder that work with different data - in this case, with images and sounds.
 
The encoder studied on the MNIST dataset, which contains 60 thousand handwritten characters, and on the Behance Artistic Media dataset, which includes about 180 thousand oil and watercolor paintings.
The neural network first performs the conversion from the image to music, it performs the reverse conversion from the received music to a new image, after which it is compared with the original one.