Skip to content
ImagineScenes
Journal

WORKFLOW NOTE 05

AI Image to Video: Start Frame, Start & End Frame, or Reference Images

A still image, a short generated video, and a second still showing three ways to guide an AI image-to-video clip

Compare Start Frame, Start & End Frame, and Reference Images. Choose by image role, check compatible models, and review the complete generated video.

Choose whether the image is a start, finish, or reference

Start Frame uses one still to guide the opening. Start & End Frame uses two images to guide the beginning and destination. Reference Images uses supported sources to guide appearance or style. Select the mode by the role your images must play.

Choose the mode before selecting a model and uploading. The editor filters compatible models and displays the relevant slots. A source image guides generation; none of these modes guarantees unchanged faces, products, typography, or every frame.

Three image roles arranged around a generated product video: an opening still, an opening and closing pair, and a set of visual references
Each mode gives uploaded images a different role: a fixed opening, a pair of endpoints, or separate creative references.

Check the two frame slots and current settings

Start & End Frame displays a start-image slot and an end-image slot for compatible models. Upload two clear compositions, then describe the transition between them. Match subject proportions and camera framing when continuity matters.

The selected model controls available duration, framing, resolution, and other options. Read upload validation and the credit estimate after choosing the settings. A screenshot or estimate from another mode does not establish the current request limits.

ImagineScenes Image to Video editor with Start and End Frame selected, empty upload slots, Seedance 2.5 selected, and an output preview
Actual ImagineScenes editor capture: selecting Start & End Frame reveals the two empty endpoint upload slots in the same generation workflow.

Use Start Frame to animate one still

Choose Start Frame for a portrait gesture, product reveal, or environmental movement from an approved image. Describe one action and a restrained camera move, plus the visual details to retain.

Review the face, product label, and geometry across the clip. The still anchors the request, but newly generated movement can cause drift. Save a useful result before testing another motion direction.

EXAMPLE DIRECTIONStart from the uploaded bottle image. Condensation gathers and moves slowly on the glass while the camera makes a controlled forward push. Preserve the bottle shape, pale green liquid, label area, and stone surface. Keep the bottle centred; do not introduce new products.
An unbranded pale green beverage bottle on stone with clear space beside it for camera movement
A strong Start Frame begins with one clear subject and a composition the team already approves. The prompt supplies the motion after this opening image.

Use Start & End Frame for a defined destination

Choose two endpoint images when the finish needs a particular composition. Describe a plausible action and camera path connecting them, such as a slow approach to the same product on the same set.

Play the full clip and inspect both endpoints as well as the transition. Abrupt changes can reflect conflicting images or an overloaded prompt. Simplify one input or instruction at a time; an ending reference is guidance rather than an exact-frame guarantee.

EXAMPLE DIRECTIONBegin with the product on pale stone in morning light and finish with the product framed against the open window. Move through a slow rightward camera slide as citrus leaves shift in a light breeze. Preserve the bottle silhouette and clear label area throughout the shot.
The same pale green product bottle shown in two matching compositions, first on stone and then beside an open window
Use two endpoint images when the closing composition is part of the approval requirement, not an accidental outcome of the generated motion.

Use Reference Images for distinct visual roles

On a compatible reference model, assign each source a purpose: character appearance, product details, scene, or visual treatment. Use the reference tags shown in the editor so instructions identify the correct upload.

Avoid contradictory faces, packaging, or styles. More references do not guarantee better consistency. Check the slot count and permitted combinations after selecting a model, then inspect which intended details appear in the output.

EXAMPLE DIRECTIONUse @image1 for the jacket, hairstyle, and silhouette. Use @image2 for the rain-lit market palette and wet pavement reflections. The character walks at a calm pace through the market while the camera tracks beside them. Keep the jacket unchanged and use a medium shot.
A fictional adult woman shown in portrait, full-body, market environment, and rain-lit market scene references to demonstrate character consistency
For character consistency, separate identity, silhouette, and setting into individual references, then name each responsibility in the prompt.

Judge the result by the chosen image role

For Start Frame, compare appearance as motion begins. For Start & End Frame, inspect both compositions and the connecting movement. For Reference Images, compare the character, product, or style details each source was meant to guide.

Every mode requires review of the full clip and any generated sound. Recheck the estimate before another submission and keep a clear comparison version. Use Video to Video when footage, rather than still images, must supply movement or timing.