How to Choose a Photo for AI Video
Prepare two portraits for a duet video: image size, lighting, face visibility, permission and common mistakes.
Give the model a clear identity
Choose a sharp photo where the face is easy to recognize. Avoid sunglasses, heavy face occlusion, extreme angles and strong blur. Even lighting makes facial features easier to interpret. Generated results can still vary; a good source photo does not guarantee an exact likeness.
Meet the upload requirements
Lite NEON Dance accepts JPG, PNG and WebP files up to 8 MB, with both dimensions at least 340 pixels. The photo checker reads your image locally to check file size, dimensions and format before you upload it to the studio. It does not assess consent or guarantee generation quality.
Match the photo to the performance
Choose People or Pets, then upload two separate, sharp photos. Each slot maps to one subject: the first appears on the left and the second on the right.
Prepare two photos for a duet
Booth Duet requires two distinct uploaded images. Use one file per person or pet rather than a crowded group photograph. Select People or Pets in the creator, then use one photo per slot. Both photos must meet the same file and dimension requirements.
Check permission and understand processing
Only upload photos you have permission to use for the intended performance. The studio sends uploaded photos to EvoLink Wan 3.0 to generate the video. The local photo checker does not send its image to the provider. Review the generation screen and avoid using another person’s image to imply a real endorsement or event.