Face Jam AI 1.5

Choose two faces and combine their styles with AI (generative neural nets).

Additional Charges:

• The watermark can be removed, and the generated image can be exported with an in-app purchase. Each generated image requires its own in-app purchase.

Some potential applications of this application include:

1. Hair Color: Set the "input" photo to the person whose hair color you wish to change. Choose a photo of a person with the desired hair color as the "style" photo. Hair color adjustments are considered shallow styles, allowing for mixing at relatively shallow depths. Mixing deeper styles will transfer profound properties from the style photo, such as face structure, gender, and head pose.
2. Baby Generator: Select photos of both parents and blend their styles. When blending deep styles, gender styles will combine; a gender slider is available to adjust this. Mixing only shallow styles will retain the gender of the "input" face.
3. Hair Transplant: This feature necessitates deeper style blending than adjusting hair color. Designate the input photo as the recipient of new hair and set the style photo to be the donor of the hair. Explore optimal depth levels for transferring the desired hair style; typically, the 2nd to last and 3rd to last layers work best in most cases.
4. Cartoons: The app can detect some cartoon faces and mix their styles with those of real individuals.

Fun fact: Many faces showcased in the app's examples were created directly within this application—these individuals are purely fictional.

Note: This application does not gather any facial data; all processing occurs locally on the device. No data is shared with third parties, and no information is stored locally. The sole method of exporting data from this app is by sharing the generated image via the button located in the lower right corner of the stylized image.

Some functionalities in this app were inspired by the following research:

1. Goodfellow, Ian, et al. "Generative adversarial nets." Advances in neural information processing systems 27 (2014).
2. He, Kaiming, et al. "Deep residual learning for image recognition." Proceedings of the IEEE conference on computer vision and pattern recognition. 2016.
3. Karras, Tero, et al. "Analyzing and improving the image quality of StyleGAN." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020.
4. He, Kaiming, et al. "Momentum contrast for unsupervised visual representation learning." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020.
5. Huang, Yuge, et al. "CurricularFace: adaptive curriculum learning loss for deep face recognition." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020.
6. Richardson, Elad, et al. "Encoding in style: a StyleGAN encoder for image-to-image translation." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021.
7. Tov, Omer, et al. "Designing an encoder for StyleGAN image manipulation." ACM Transactions on Graphics (TOG) 40.4 (2021): 1-14.