Video synthesis within a messaging system

A system comprising:

• a processor;
• and
• a memory storing instructions that, when executed by the processor, configure the system to perform operations comprising:
• access a primary generative adversarial network (GAN) comprising a pre-trained image generator, a motion generator comprising a plurality of neural networks, and a video discriminator;
• generate an updated GAN based on the primary GAN, by performing operations comprising
• identify input data of the updated GAN, the input data comprising an initial latent code and a motion domain dataset, and
• train the motion generator based on the input data;
• and
• generate a synthesized video based on the primary GAN and the input data.

A non-transitory computer-readable storage medium, the computer-readable storage medium including instructions that when executed by a computer, cause the computer to perform operations comprising:

• access a primary generative adversarial network (GAN) comprising a pre-trained image generator, a motion generator comprising a plurality of neural networks, and a video discriminator;
• generate an updated GAN based on the primary GAN, by performing operations comprising
• identify input data of the updated GAN, the input data comprising an initial latent code and a motion domain dataset, and
• train the motion generator based on the input data;
• and
• generate a synthesized video based on the primary GAN and the input data.