Trash bags: first-last empty can, then the seated liner
Two photos. Interpolate. The carton face stays a still with overlay count.
Two photos. Interpolate. The carton face stays a still with overlay count.
A trash-bag ad that sells is an empty can, then the same can with the liner seated — drawstring, cuff, the colour of this SKU. It is not a talking founder over a kitchen, and it is not a carton that typesets "90 count" in generated glyphs. You already own both stills of the can. You already own the carton face. Do not prompt a model to invent either.
/for/cleaning-services is a crew page: mop, truck, a named cleaner if a human is the offer. This post is the CPG liner. Empty can to seated bag is a product turn, not a janitorial walkthrough.
Frame A is empty; frame B is the seated liner
Photograph the can empty — same lens height, same tile, same lid out of frame or locked. That is the first frame. Photograph the same can with the liner seated: cuff over the rim, the print if this SKU has a print, the drawstring where it actually sits. That is the last frame.
First-last frame takes those two stills and generates the motion between them. Ordinary image-to-video knows where it begins and improvises the rest. First-last has to land. The arc of the shot is decided by you.
The two frames have to be connectable in the time available. Same can, same lighting family, same camera. A close-up of a box flap morphing into a wide kitchen is a warp, not a seat. Keep the can. Keep the rim.
Do not use a last frame that is a different room. The model will connect them. Do not treat a loop (same still in both slots) as the liner ad. A loop is idle motion on one plate. This offer needs two states.
Interpolation is the seat, not a kitchen film
Create a first-and-last-frame transition video is the named agent job: two photos in, one transition out. The agent passes them as first and last frame and picks a first-last-capable model. Cost is priced per model, shown before you confirm. This is not the plain "turn one photo into a video" capability.
Flux 3 First Last Frame to Video is the catalog row for that interpolation: start frame and end frame, native audio, 720p or 1080p, durations from 5 seconds through 20. The listed rate is 9 credits per second. Audio is on — write the mix (a quiet rustle, a bag unfolding) or write silence. A silent brief still gets a soundtrack you did not choose.
Prompt the travel, not the objects. The stills already contain the empty drum and the seated liner. Name the mechanism: hold, then the bag drops and cuffs; camera locked; no new cabinets. If the endpoints share geometry, the morph reads as a seat. If they do not, you get a mushy crossfade you could have made in an editor without a video bill.
Duration should match the morph, not a Reels quota. A short travel between two honest can photos reads as proof. A long travel invents a second kitchen in the middle: the middle is where a generator cheats.
Inspect the midpoint. Ends are held by construction. If the middle grows a different print, a phantom box, or a can that changes diameter, reroll the interpolation. Do not regenerate both stills "to match the motion." The stills are the contract. The motion is the cheap middle.
The carton face is a still, not the interpolate
The count, the scent line, the lockup on the box — that is a pack plate. It is not frame B of the can. Do not first-last the empty can to the carton and call the middle "our liner." Two different objects is a morph. Two stills of the same can is a seat.
Lock the carton as its own still. Animate it only if the job is a pack orbit, with a second approved still of the same box. Most weeks you do not need that orbit. You need the can pair as the proof clip, and the carton as a hold or a cutaway.
Text-to-video "90-count white kitchen bags seating themselves" rebuilds the box and the kitchen together. Close is how you get a count the carton does not print and a drawstring the SKU does not use. Make the stills. Then interpolate the can. Cut to the carton.
If you needed a founder line about toughness, cut it after the seated liner. The can shot does not have to talk.
Overlay the count after picture lock
"90 count," the scent name, a SKU string — those are authored lines. They do not belong in Flux glyphs and they do not belong in a carton the model restocked. Add a text overlay burns one line you wrote, whole clip, top / center / bottom. It does not transcribe. A mute can plate has nothing to caption.
Do not prompt "print 90 COUNT on the box" into the interpolate. Prompted type is a miss you cannot edit against the die-cut. Keep the carton photograph honest. Burn the real string on the file you are keeping. Crop to the ratio you will ship before you overlay.
Muted autoplay still needs the count on screen if the carton face is not in frame. Overlay does that. There is no speech on an empty can.
The test: start picture, end picture, travel without inventing a kitchen? Yes → first-last on the can. The carton stays a still. The count stays type you typed.
FAQ
Can I animate one photo of the empty can and hope the bag appears?
You can. The ending is then the model's guess. The liner ad is the seated cuff. Pin the seated can as the last frame or you are not selling the seat.
Why Flux 3 instead of a talking-head kitchen?
Because the product is interpolation between two approved stills, not a mouth. Flux 3 first-last is 5–20 seconds, 720p/1080p, native audio. A presenter row will talk over a can it invented.
Do the two photos have to be shot on the same day?
They have to share geometry. Same can, same lens family, same lighting neighbourhood. A noon still into a night still is a time-lapse you should prompt as one. A different kitchen is a morph.
Should the carton interpolate with the can so shoppers see the count?
No. The carton is a different object. Hold the carton still. Overlay the count. Interpolating can-to-box is a warp, not a listing.