Upload clear reference photos of your chosen characters, then use Media.io Image to Video with a prompt describing the small knight, tall guardian, misty field, and sequential sword exchange. For greater control, first prepare a consistent knight scene in Image to Image. The exact workflow depends on your selected AI model.

Try: “Two knights stand on a misty green meadow at dawn. A tiny child-sized knight in a red cape struggles to lift a heavy sword. A taller adult guardian knight approaches slowly, kneels on one knee, gently receives the sword, and offers a small bouquet of flowers. The tiny knight looks up in surprise. Warm emotional storytelling, cinematic slow camera push, realistic armor, stable faces and hands, consistent scale, vertical 9:16.” Complex object handoffs may need shorter generations.

Yes. Use a parent-and-child pair to show protection and growth, or two friends for a supportive gym-bro tribute. You can also replace the flowers with a toy. Use only images you are authorized to edit, especially when featuring children, and keep the characters’ scale and actions clearly described.

Keep the camera movement slow, show both knights’ hands, and describe each action in order: the little knight lifts the sword, the guardian kneels, accepts it, and offers flowers or a toy. Avoid simultaneous fast gestures. Review the generated video for extra fingers, disappearing objects, and unstable armor; results vary by model.