Make a cafe drink or a dish look delicious. Use it for restaurant intros and cafe-tour vlogs.
In Shot 1, the camera slides slowly sideways along the counter. The blurred cup in front, the barista in the middle and the lights above pass at different speeds, which brings out the depth of the cafe.
To recreate it: Give only one sideways direction like 'slow truck right', and describe the objects in the foreground, middle and background separately in the prompt.
Move very close, as in the espresso extraction in Shot 2 and the cheesecake cross-section in Shot 4, to show the crema and crumbs large. When texture is visible, the taste can be imagined.
To recreate it: AI handles macro shots without people fairly well, so write 'macro shot, 100mm macro lens' together with a fixed camera.
In Shot 2, strong light from behind the coffee makes the liquid glow amber. In Shot 5, it lights the outline of the head of a person with their back to the window. Light from behind brings out steam, foam and gloss.
To recreate it: Put the light position and the background color in one sentence, like 'strong backlight against a black background'.
In Shot 3, look straight down at a latte on a marble table and catch the moment the milk draws a heart. From above, the latte art shows without distortion.
To recreate it: The pattern often comes out messy, so generate several 480p drafts and pick the cleanest heart.
In Shot 2, stretch the brief moment when the espresso drops and the crema swirls. Showing an instant for longer is what makes people want to eat it.
To recreate it: Put 'slow motion' in the prompt from the start, and don't force-stretch it in editing.
Join extraction, pouring and cutting inserts in short 0.5-1s pieces and cut to the music beat (beat sync). Short actions lock onto the beat and create the rhythm of a busy cafe.
To recreate it: Generate AI clips at 4-5s and use only the cleanest moment, under 1s.
At the end, bring the colors that vary from shot to shot because of cafe lights, window light and backlight into one warm tone. With unified color, the many shots read as one cafe.
To recreate it: Don't over-specify color at generation, and apply the same grade values to all shots at once in editing.
Cinematic multi-shot sequence, 11 seconds total. Shot 1 (3s): slow truck right along a cozy cafe counter, blurred cups in the foreground, a barista in the midground, warm pendant lamps; Shot 2 (2s): static macro of espresso streaming into a small glass in slow motion, strong amber backlight; Shot 3 (2s): static top-down flat lay of a heart pattern poured into a latte on a marble table; Shot 4 (2s): close-up insert of a fork cutting a slice of cheesecake, shallow depth of field; Shot 5 (2s): medium close-up of a young man by the window taking a bite and smiling, soft rim light. Consistent warm cafe interior, soft window light, 35mm film grain, indulgent happy mood.
Very close shots and light coming from behind bring out the steam, foam, and shine, so the food looks delicious. Link short actions like pouring and cutting on the beat, and you get the rhythm of a busy cafe.
Slow truck right along a cozy cafe counter, blurred cups in the foreground, a barista working in the midground, warm pendant lamps glowing above, 35mm lens, shallow depth of field, soft 35mm film grain, inviting afternoon mood.
Macro shot of dark espresso streaming from a portafilter into a small glass cup in slow motion, crema swirling, static camera, strong backlight making the coffee glow amber against a black background, rich warm tones, fine film grain, satisfying mood.
Top-down flat lay of a latte on a marble table as steamed milk is poured into a heart pattern, static overhead camera, soft window light from the left, a croissant and spoon neatly placed, warm creamy colors, gentle film grain, calm cafe mood.
Close-up insert of a fork cutting into a soft slice of cheesecake, crumbs falling onto the plate, static camera, 100mm macro lens with shallow depth of field, warm side light from a window, glossy texture visible, 35mm film look, indulgent mood.
Medium close-up of a young man at a window seat taking a bite of cake and smiling with closed eyes, static camera, soft daylight from the window behind him with a gentle rim light, blurred cafe background, warm film tones, happy mood.
Show the whole cafe first to say where we are. Then cut the brewing, pouring, and cutting inserts quickly, 0.5–1 second each (fast cutting), in sync with the beat of the music. Turn up the sound of the coffee machine and the fork, and turn the music down. Hold only the last reaction shot long, and color grade everything to a warm tone.
Generate each shot separately, and settle the plate layout and light direction first in the start frame image. Liquids and steam come out well, but fingers holding a cup or a fork are often smeared. Make several 480p drafts of those shots, and of shots where the shape matters, like latte art, and pick the best. Slow motion looks choppy if you stretch it in editing, so put "slow motion" in the prompt and generate it slow from the start.