Lesson Content

Why animate from an image.

Starting from a still image anchors the generation to a specific visual – useful when you have a target visual that needs to be animated (a rendered illustration, a photograph, or an AI-generated still) and want the video to maintain that visual's look while adding motion. Pure text-to-video would require recreating the visual through prompting, which may not precisely match the existing asset.

Preparing the source image.

Higher quality source images produce better animation results. Images with clear subjects against relatively distinct backgrounds animate more predictably than cluttered or visually complex images. If the source image has significant noise, compression artifacts, or very low resolution, these characteristics may carry through to the animated output.

Adding motion direction.

After uploading the image, add a text prompt describing how the image should animate – "gentle wind movement through the flowers," "slow camera pull-back," "character's hair moving softly," "waves rolling in the background." The motion prompt guides animation behavior without overriding the source visual's appearance.

Evaluating animated results.

Check: does the motion match the intended direction? Is the source image's visual maintained without excessive distortion? Are any specific elements being distorted that should remain stable? If faces or fine text are distorting significantly, the motion intensity may need to be reduced.