UV print on objects: the object still is the brief
I2V the mug/phone you photographed.
I2V the mug or phone you photographed. The product is the object that left the printer: the ink on that tumbler, the registration on that case, the colour you already signed off. A text-to-video "UV printed mug on marble" is a second mug. Labels, bleeds, and the print you actually sold will drift. The still is the brief.
Photograph the piece. Animate that file. Do not invent a better print.
The print is already the design
UV work is judged on the object in hand. A buyer who ordered a logo on a 12oz tumbler is looking for their mark on that curve. A model asked to imagine the job will re-typeset the logo, shift the wrap, and pick a nicer ceramic. That is a concept render. It is not a proof of the job you ran.
Shoot even light, plain background, front of pack, a three-quarter, the print close enough to read. That is the highest-leverage ten minutes of the ad. If the still is wrong — dust, a bad wrap, a colour you would not ship — stop. Every second of video from a bad still is waste.
Do not regenerate the artwork onto a generated mug. Prefer the photograph of the finished piece. The printer already did the hard part.
Motion from the still, not a new object
Seedance 2.5 Image to Video is the still-to-motion row: one still into a continuous clip at up to 720p, durations from four seconds to thirty, native audio with the picture. You can add a second image as a last frame when the move has a destination — a three-quarter that must land on the back of the wrap, a push-in that must end on the mark.
Thirty seconds in one pass is why this family keeps the pack-shot slot. Stitching four short cinematic clips around a tumbler is how the print drifts. One pass from a locked frame is how it does not.
Image-to-video is the tool shape. Prompt only the motion: slow orbit, table turn, light sliding across the varnish. Do not re-describe the logo. The model can already see it. 720p is the ceiling on this endpoint. If you need a 2K plate for a grade, you are on a different row. If you need the print to remain the print, you are here.
When reference-to-video is the extra lock
A single hero still is usually enough. When the object has to hold across a scene the still does not contain — a desk, a hand, weather you will not shoot — use reference-to-video. AI product video is that roster: two or three clean angles tell the model what must not change (shape, colour, the printed face) while the rest of the frame is open.
R2V is many files, addressed in one generate. I2V is one still, or a first and a last. A mug on a sweep table does not need fifty reference slots. It needs the frame you already signed off, moving. Zoom to 100% on the keep and read the small print. Fine text is the first thing a generate loses.
Proof the object, then ship the move
Pause on the label. If the wrap is not your wrap, the take is not an ad. Caption after, if you need a SKU line or a "printed this week" stamp — type it; do not ask the video model to letter the lid. The still remains the contract for as long as that SKU is on the site.
FAQ
Can I text-to-video the mug and skip the photo?
You can. You will re-decide the print every take. Colour, wrap, and the actual logo are the expensive failure. The still is the cheap place to lock them. I2V is how you stop paying video credits to discover the tumbler changed.
Do I need a last frame as well as a first?
Only when the shot has a destination that must match a second approved still — the back of the wrap, a mark you already photographed. A single hero still with a slow orbit is the usual job.
Seedance I2V or the product-video tool?
Same object, different input. One approved still: Seedance 2.5 Image to Video. Several angles that must hold in a new scene: the product-video generator's reference-to-video roster. Do not send a blank prompt and a moodboard and call it a pack shot.
Is 720p enough for a print shop reel?
For feed and most web clips, yes. For a 2K landing-page plate, pick a 2K generate and still start from a locked still if you can. Resolution is a delivery choice. Identity is the I2V choice.