5 ms·

Interactive demo here: https://www.kaggle.com/summitkwan/tl-gan-demo https://www.kaggle.com/summitkwan/tl-gan-demo

simlevesque 8y ago
HN effect in full effect
transpy 8y ago
It's very interesting and kinda creepy. The controls behave randomly, if you decrease "age" it might start changing skin color, etc.
- e_ameisen 8y ago
  Actually, you can click on the name of the attribute to "lock" it. That way you can change one without changing the other!
  - transpy 8y ago
    Excellent
- subcosmos 8y ago
  Oh yeah! I wrote a similar algorithm and constantly suffered similar problems. Constraining the latent space can be a major pain. Its hard to visualize but I think a lot of latent features may exist in non-continuous latent "pockets", meaning that the desired direction of the latent vector is dependent on your position in space.
  - transpy 8y ago
    I´m having a blast changing all the parameters. Really uncanney valley feelings. Some women come up very attractive, looking like Jessica Alba. How can I react to a face of a person that don't even exist? On a second thought, I don´t mean that the controls behave randomly. I understand they affect related parts of the system. If you increase "baldness" on a woman's face, it will obviously increase the "male" factor and the "gray hair" factor. I understand that this faces are being generated from a continuous space. Fascinating.
    - modeless 8y ago
      Looking like Jessica Alba is not a coincidence. This model was trained on CelebA, a dataset of Hollywood celebrity face photos. Jessica Alba is almost certainly in the training set.
subcosmos 8y ago
Very very rad! I've got a similar algorithm https://bit.ly/2ELVG50 https://bit.ly/2ELVG50 To use, upload a photo with the #ageme command. This is beta though, so will take a while to return. The other thing this bot does (the #showage command) runs instantly. Example output: https://bit.ly/2PXKTWu https://bit.ly/2PXKTWu Your model seems to produce much higher quality output. I think I played a penalty by putting the users face into latent space, which is a difficult optimization to perform.
- VikingCoder 8y ago
  (It's not mine, I just found the link in the article)
IshKebab 8y ago
It would be cool to integrate a face tracker to measure features, then you could learn the latent vectors for things like eye position and then you could drag the eyes around.

HN effect in full effect

It's very interesting and kinda creepy. The controls behave randomly, if you decrease "age" it might start changing skin color, etc.

e_ameisen 8y ago
Actually, you can click on the name of the attribute to "lock" it. That way you can change one without changing the other!
- transpy 8y ago
  Excellent
subcosmos 8y ago
Oh yeah! I wrote a similar algorithm and constantly suffered similar problems. Constraining the latent space can be a major pain. Its hard to visualize but I think a lot of latent features may exist in non-continuous latent "pockets", meaning that the desired direction of the latent vector is dependent on your position in space.
- transpy 8y ago
  I´m having a blast changing all the parameters. Really uncanney valley feelings. Some women come up very attractive, looking like Jessica Alba. How can I react to a face of a person that don't even exist? On a second thought, I don´t mean that the controls behave randomly. I understand they affect related parts of the system. If you increase "baldness" on a woman's face, it will obviously increase the "male" factor and the "gray hair" factor. I understand that this faces are being generated from a continuous space. Fascinating.
  - modeless 8y ago
    Looking like Jessica Alba is not a coincidence. This model was trained on CelebA, a dataset of Hollywood celebrity face photos. Jessica Alba is almost certainly in the training set.

subcosmos 8y ago

Very very rad! I've got a similar algorithm https://bit.ly/2ELVG50 https://bit.ly/2ELVG50 To use, upload a photo with the #ageme command. This is beta though, so will take a while to return. The other thing this bot does (the #showage command) runs instantly. Example output: https://bit.ly/2PXKTWu https://bit.ly/2PXKTWu Your model seems to produce much higher quality output. I think I played a penalty by putting the users face into latent space, which is a difficult optimization to perform.

VikingCoder 8y ago
(It's not mine, I just found the link in the article)

IshKebab 8y ago

It would be cool to integrate a face tracker to measure features, then you could learn the latent vectors for things like eye position and then you could drag the eyes around.