I started modelling with reference pictures, using two photos as the front and side view to keep it as proportional as possible, the ear was one of the most difficult things to model as well as keeping the mesh uniform and shape the face as close to reality as possible. I couldn't find a good solution for the hair, because it just becomes round with the subdivision surface. After that I marked some seams to divide the model and unwrap it, the reference photos didn't have the best light or quality, I spent a lot of time doing colour correction. In the end I've recognized some situations that should be corrected earlier in the process, in order to save up time and have a better model. I made an animation and edited in Blender video editor.