Rainbow Bridge AI Video Prompt: Meet Your Dog Again

Photorealistic character reference sheet of the exact same dog from @Image1. Studio photography on a plain seamless pure white background, soft even light, soft natural shadow under the feet only, sharp focus, real fur texture.
Keep the dog from @Image1 exactly: the same breed, size, body shape, coat colours and pattern, face markings, ears, eye colour, nose and tail. Do not change, clean up or beautify anything. No collar, no harness, no clothes.
Layout: one clean sheet with three rows, no frames, no text.
Top row, five full-body views at the same scale, feet on one baseline: standing facing the camera with mouth slightly open; left side profile; back view showing the tail; right side profile; sitting facing the camera.
Middle row: a large head portrait from the front; a three-quarter head view with mouth open and tongue out; a side profile of the head; a top-down view of the head looking up; a very close crop of the face from eyes to nose.
Bottom row, square detail tiles: one eye in extreme close-up with a soft natural catchlight; the nose; the inside of one ear; the underside of one paw with the pads; then the dog lying down with its head on its paws; then the dog sitting seen from behind.
Every panel shows the same single dog. One dog per panel.
Avoid: text, labels, numbers, logos, watermark, extra dogs, other animals, people, props, coloured background, the background from @Image1, cartoon, illustration, 3D render.

Ultra-photorealistic character reference sheet of the exact same person from @Image1. Studio photography on a plain seamless light-grey background, soft even daylight, natural soft shadows, 50mm lens look, sharp focus, real skin texture with pores and freckles, no beauty filter.
IDENTITY: keep the person from @Image1 exactly: the same face, bone structure, eye shape and eye colour, eyebrows, nose, lips, skin tone, freckles and marks, age, hair colour, hair length and hairstyle. Do not beautify or change the face.
CLOTHING: a relaxed beige linen button-up shirt with a chest pocket and sleeves rolled to the elbow, worn open over a plain white top; slim mid-blue jeans; clean white leather trainers. No jewellery, no bag, no hat.
LAYOUT, small black labels in the top-left corner of each panel:
Top row, three full-body panels at the same scale, head to shoes, feet on one baseline, relaxed standing pose: “1. Front View”, “2. Side View” (facing right), “3. Back View”.
Middle row, two large panels: “4. Front Close-Up”, head and shoulders turned slightly, soft natural smile; “5. Top-Down View”, seen from above, looking down.
Bottom strip, five equal detail tiles with a small title above each: “Hair Detail”, “Eye Detail”, “Shirt Fabric Detail”, “Jeans Fabric Detail”, “Trainers Detail”.
Thin white gutters between all panels.
Every panel shows the same single person in the same clothes, the same hair and the same face.
Avoid: extra people, other outfits, accessories that are not listed, face changes, different hair colour, the background from @Image1, cartoon, illustration, 3D render, watermark.

[References]
The dog corresponds to @Image1. Use only its breed, coat colour and pattern, face markings, ears, eye colour and body size. Do not use the studio background, the multiple poses or the close-up detail tiles. Show only one dog, with no collar.
The woman corresponds to @Image2. Use only her face, hairstyle, hair colour, eye colour, skin details and the exact clothing and shoes she wears in the reference. Do not use the studio background, the text labels, captions or the detail tiles. Show only one woman.
All views in @Image1 show the same single dog. All views in @Image2 show the same single woman.
[LOOK — LOCKED, APPLIES TO EVERY SHOT]
ARRI Alexa Mini LF, Cooke S4/i primes, Black Pro-Mist 1/4, spherical, T2 — large-format shallow focus with the background melting into soft haze, warm natural skin, creamy round bokeh, gentle bloom around highlights, fine organic grain, soft highlight roll-off, lifted pale shadows, low contrast.
Heavenly Pastel grade: luminous emerald grass as the hero tone, warm golden light and pearl-white sky, soft peach in the shadows.
Steadicam glide with no footfall bounce, 50 fps slow motion on the emotional beats. Natural, restrained acting: calm faces, small real micro-expressions, no exaggerated grimaces. NO MUSIC.
[Scene]
Paradise, but photoreal: an endless soft emerald meadow full of tiny white and pale-yellow wildflowers, gentle rolling hills fading into haze, a pastel sky with glowing pearl-white clouds and golden light rays falling through them. Pollen and small petals float slowly in the air.
[Stage 1 — 0-3 seconds]
Initial state: extreme close-up of the dog’s eye: a natural dark-brown dog eye with a soft moist surface and one small, soft catchlight. No mirror-like reflections, no landscape reflected in the eye, no glassy shine.
Camera: 100mm macro, very slow push-in toward the pupil, focus on the eyelashes.
Primary event: the dog blinks slowly once, calm and gentle.
End state: the dog’s eye is open, looking forward.
[Stage 2 — 3-6 seconds]
Initial state: extreme close-up of the woman’s eye, strands of her hair at the edge of frame.
Camera: 100mm macro, very slow push-in, focus stays on the lower lashes as the tear forms.
Primary event: her eye slowly becomes wet, then one tear spills over and rolls down her cheek. Her face stays still and calm.
End state: a single tear line is visible on her cheek.
[Stage 3 — 6-9 seconds]
Continue from the previous stage: same tear on her cheek, same outfit.
Camera: 85mm, slow pull-out from a close-up to a medium close-up, the blurred meadow opening up behind her.
Primary event: soft wind in her hair. She notices something in the distance: her eyes widen slightly, then a soft, quiet smile slowly appears, lips closed. Calm and tender, not dramatic.
End state: the woman stands in the meadow with a gentle smile and a tear on her cheek, looking past the camera.
[Stage 4 — 9-13 seconds]
Camera: 24mm, low angle at grass level, the camera tracks backward in front of the dog, keeping its face centred.
Primary event: the dog runs toward the camera through the flowers at full speed, ears back, tongue out, in slow motion. Petals and pollen burst up behind its paws.
End state: the dog is a few metres from the woman, still running.
[Stage 5 — 13-16 seconds]
Continue from the previous stage: same meadow, same light, one dog, one woman.
Camera: 35mm, medium-wide side view at her kneeling height, slow push-in as the dog reaches her.
Primary event: the woman kneels down and opens her arms wide with a warm, open smile. The dog leaps into her chest and its weight knocks her gently backwards onto the soft grass. Flowers and petals fly up around them.
End state: the woman lies on her back in the flowers, the dog standing over her with its front paws on her shoulders.
[Stage 6 — 16-20 seconds]
Camera: 35mm, top-down overhead view, a slow crane up that ends in a wide shot.
Primary event: the dog happily licks her face again and again, tail wagging. The woman smiles softly with her eyes closed and wraps both arms around the dog’s neck.
End state: wide overhead view of the woman and the dog hugging in the middle of the flower meadow, bathed in golden light.
[Maintain Consistency]
Exactly one dog and one woman throughout. Keep the dog’s breed, coat pattern and size, and the woman’s face, hair, eye colour and outfit identical to the references in every shot. Keep the same meadow, sky and light direction.
[Audio]
NO MUSIC. Sound effects only, no dialogue, no subtitles.
Stage 1-3: <soft wind over grass>, <distant birdsong>, <her slow, soft breathing>.
Stage 4: <fast dog paws thudding on grass>, <happy panting>, <grass and flowers rustling>.
Stage 5: <her short soft laugh>, <soft thump as she falls onto the grass>.
Stage 6: <dog licking and snuffling>, <her quiet happy laugh>, <gentle wind>.