Reference assets

1 image
  • Drawer Organizer Explainer product reference photo1
Format
9:16
Length
30s
Tool
image to video

These are the assets and settings this result used. The example video is the result, not an input.

AI Video Generator for Brands, Agencies, and Creators

Drawer Organizer Explainer

Format
9:16
Length
30s
Tool
image to video

Opens the workspace with this template already selected. You can change the product, format and length.

Prompt

Overall: 30-second vertical TikTok Shop shopping video, handheld phone-shot UGC style, kitchen light, whole frame crisp. One continuous take at one kitchen counter: she turns one crammed cutlery drawer into an ordered one end to end, and ends smiling beside the finished drawer. Only one person is ever visible. Product: Image 1 is this video''s product: a two-piece bamboo cutlery tray that stays exactly as shown in Image 1 in wood tone, division and proportion, with no text or logo anywhere on it. Both pieces stay rigid and flat at all times - they never bend, warp or curve. Seating and loading follow the shots below with real contact and flat seating. Creator: A woman in her 30s in a plain olive tee with no logo, bare forearms, no jewelry; speaks natural American English like sharing with a friend, not like an announcer. She stays at the same counter for the whole video. Shots: Shot [0-3s] medium shot: she pulls the crammed cutlery drawer fully open, handles crossing everywhere; end state: both hands ready. She says: "This bamboo tray fixed my chaos drawer in minutes." Shot [3-9s] close-up: she sweeps the loose utensils into a pile on the counter beside the drawer, then drops the two bamboo trays into the drawer and presses them flat - seated flush; end state: the trays level in the drawer, the pile beside them. She says: "Lift the old jumble out, drop the tray in, and it locks flat." Shot [9-18s] close-up: forks laid in row one, knives in row two, spoons in row three, every handle the same direction; the drawer glides shut and opens again to show the rows; end state: closed. She says: "Forks land in one row, knives in the next, spoons right after - every handle sits the same way, and the drawer finally closes smooth." Shot [18-25s] medium shot at the same counter: she pulls the drawer open and takes out a fork, then a spoon, each in one motion without looking; end state: fork and spoon in hand. She says: "Now I grab what I need in one motion, and cooking prep takes half the time now." Shot [25-30s] the camera holds: the open drawer with its lined-up rows stays beside her and she smiles to the camera; end state: still. She says: "One tray, and chaos is gone." Sound: Her voice carries only the quoted lines in a friendly natural tone; natural sound: utensil clinks, the tray drop, the drawer glide. No background music. She speaks only the quoted English sentences and stops when done. Constraints: Physical rules: one person only - the same woman in the same olive tee with the same bare forearms in every shot, never appearing twice in the frame; one camera side and one counter for the whole video - no stove, no pot, no cooking; the two bamboo trays stay rigid and flat and never bend or warp, the utensil pile stays on the counter, and every utensil lands with a real point of contact; her clothing stays plain and logo-free. Negative: background music, subtitles, on-screen text, text on packaging, garbled letters, watermark, logo, extra fingers, deformed hands, extra people or hands, duplicate selves, mirror or mirror selfie, product melting or deforming, stains or burn marks, background bokeh, depth-of-field blur, blurry background, two identical people, duplicate person, the same person twice, duplicated objects, object morphing, warping or bending rigid objects, floating objects, spilled water, a stove or cooking pot

The workspace opens with this prompt so you can change the shot, product, format and length before you generate.