Introducing 'FLUX Portrait Trainer,' an AI learning tool for generating images specializing in portrait photography
The development team behind the image generation AI FLUX has released a new trainer called FLUX Portrait Trainer, which is designed to generate high-quality portraits. Features include finer details, eye highlights, and better representation of various face sizes.
Train Flux LoRAs For Portraits | Training | AI Playground | fal.ai
Introducing the FLUX Portrait Trainer
https://blog.fal.ai/introducing-the-flux-portrait-trainer/
Below is a sample image generated using the same model from FLUX Portrait Trainer. Fine details such as individual hairs, highlights in the eyes, and skin texture are expressed with high quality, and the development team says that 'the picture looks like it's jumping out of the screen.'
The image below was generated based on the dataset provided by CheeseAI, with the prompt 'kptz woman in a pitch black room. I single shaft of volumetric light slants across her right eye.' The image includes the instruction 'pitch black room,' but the dataset only contained photos taken during the day. The image on the left was generated by a model trained with FLUX Portrait Trainer, and the image was output as intended, but the image on the right was generated by a model trained with a general-purpose trainer, and could not be 'pitch black.'
Similarly, the following images were generated under the same seed and parameter conditions for the prompt 'kptz old woman.' The image generated by the model trained with FLUX Portrait Trainer on the left is the image requested by the prompt, while the image generated by the model trained with the generic trainer on the right is an image that matches the data used for training, ignoring the prompt.
The FLUX Portrait Trainer is a complement to the general-purpose trainer, and the development team explains that a different trainer may work better for some applications.
Related Posts:
in Web Service, Posted by logc_nt