100 more photo effects and video prompts for Higgsfield
The short names are labels for the ideas. They are not verified built-in slash commands. Paste the full prompt so the image or video tool receives the actual direction, including what should stay unchanged.
Set up the generation route
- Open the Plugins area in ChatGPT, find Higgsfield and choose the available Add or Install option. If you do not see that area, start from Higgsfield's connection page and follow the instructions available for your account.
- Connect or sign in to your Higgsfield account when prompted, then start a new conversation. You can find the service at Higgsfield (my affiliate link). I may earn a commission if you purchase through this link, at no extra cost to you.
- Attach your source photograph and say whether you want an image or video. For a video, explicitly ask to generate through Higgsfield. For a still, choose the image-generation tool available in your chat, or the appropriate Higgsfield image tool if connected.
- Paste one full prompt. Check the selected model, supported duration, aspect ratio and credit cost before generation. A text description of a shot is not the generated clip, so confirm that the generation tool was actually used.
The setup labels can vary by account and rollout. OpenAI's plugin guidance and Higgsfield's connection instructions are the current setup references. A particular ChatGPT model choice is not listed as a requirement in Higgsfield's setup instructions. Product access and generation credits are separate from the wording of your prompt, and connecting a tool does not make every generation free.
Choose a source that supports the shot
Use a sharp image with an unobstructed subject, enough space around the edges and lighting that makes the important details visible. For a product, include the label and any surfaces you want to show. For a person, use a photograph you have permission to animate. Two or three clear references are more useful than asking a model to guess an unseen face angle, clasp or internal mechanism.
A request for an orbit or exploded arrangement produces a rendered visual. It does not automatically create a usable 3D asset, reconstruct accurate engineering or prove how the product is assembled. If you need exact component relationships, supply separate part photographs or a reliable assembly reference.
The prompts below describe the intended result, not a guarantee that a model will preserve every letter or movement. Start with one small generation. Check the subject, label, hands and contact points before extending the shot or spending credits on variations.
The four effects from the reel
Product explosion
Output: a still image showing a conceptual exploded arrangement. Supply: the complete product plus separate photographs or a reliable reference of the parts you want separated. You can animate the approved image afterwards if you want motion.
Using the attached complete-product photo and component references, create one vertical still image of the product in an orderly exploded arrangement against a clean charcoal background. Separate only the documented components by small, clearly visible gaps, preserving their relative orientation, proportions, materials and existing label. Use soft studio lighting and restrained shadows to make the arrangement readable. Do not invent internal mechanisms, hidden layers or included accessories. If the photographs do not establish separable parts, keep the product intact and arrange only the supplied accessories around it. This is a conceptual rendered illustration, not an engineering reconstruction or a usable 3D model. No added text.
If it adds unfamiliar parts, simplify the arrangement and supply individual component images. A single exterior photo is not enough evidence for an accurate internal build. For video, animate the approved arrangement with a small separation movement and the camera fixed.
Man POV
Output: a first-person video from the viewpoint of the person in your supplied photo, with a smart-glasses-style camera position. Supply: the scene photo and any extra reference showing what is in front of that person.
Using the attached photograph, generate a short vertical video through Higgsfield from the first-person viewpoint of the person shown, as if a small camera were mounted at their eye level on smart glasses. Carry forward the photographed setting, activity and visible clothing rather than moving them to an unrelated location. Show the space directly ahead with a subtle natural head movement and one simple action supported by the reference. Include hands only if that action needs them, matching supplied references. Do not show an invented face, add extra fingers or fabricate readable signs. If the person's forward view is not established, ask for that reference before generating. No cuts, captions or audio.
POV describes where the camera is. A photograph of someone does not necessarily show what they can see, so add a forward-facing scene reference when the view is missing. Keep the first action small enough to inspect.
Orbit shot
Output: a full 360-degree camera orbit around a stationary subject. Supply: clear front, side and rear photographs, ideally with consistent lighting and the intended background.
Using the attached front, side and rear reference photographs, generate a vertical video through Higgsfield in which the camera completes one smooth full 360-degree orbit around the stationary subject and returns to its starting view. Choose a supported duration long enough to keep the movement readable. Maintain a consistent camera height, subject scale, lighting and coherent background throughout. Keep the subject anchored in place without rotating it independently. Match the supplied surfaces, proportions and identifying details as each angle appears. Do not invent hidden product mechanisms or additional accessories. Hold briefly at the start and end. Produce one continuous rendered shot without cuts, captions or audio.
This is a generated video, not a recovered 3D model. Even with several references, views between them are inferred and may contain errors. If you only have a front photo, use a smaller 20-degree arc as a fallback and keep it from revealing unsupported rear details. Check every side of a full orbit, not only its first frame.
Slow motion
Output: one simple action played slowly. Supply: a photo that clearly shows the object and the surface where the action will happen.
Using the attached product photo, generate a short vertical video through Higgsfield of one narrow strip of lightweight tissue paper drifting slowly down beside the stationary product. Let it fold naturally in the air, touch the tabletop and settle. Keep the product completely unchanged and the camera locked at a gentle three-quarter angle. Use soft side lighting that makes the paper's movement visible without covering the label. Show a smooth slow-motion appearance with plausible gravity and contact, rather than suspending everything indefinitely. Do not add liquid, flying product parts or multiple competing actions. Finish with a brief still hold, without captions or audio.
Slow motion works best when the viewer can follow one action. If the model struggles, use a shorter drop and keep the camera still.
How to use the 100 additional prompts
These are 100 additional image and video prompts, separate from the four demos: 55 video prompts and 45 still-image prompts. Each entry names an image or video output. The categories help you find a job quickly: portraits, optical effects, constructed product sets, hands, food, fashion, spaces, layouts and transitions.
For entries that need a specific subject, supply it. The zipper prompt needs a visible zipper; the two-scene transition needs two photographs. If the input does not fit, choose another prompt rather than asking the tool to invent the missing evidence. Portrait treatments and mock layouts are editorial visuals, not records of real print publication or product performance.
For transitions 91–100, supply the two or three photographs named in the prompt and choose a model that accepts that reference workflow. Start-frame, end-frame and multi-image controls vary by model. If your model cannot use the required inputs, create separate shots and assemble the transition in your video editor instead.
For video, ask Higgsfield to use a short duration supported by your selected model. For still images, keep the request to one finished image. Most prompts use vertical framing; two editorial layouts explicitly use horizontal framing. You can change the requested framing to square or landscape when your destination requires it. Add only one revision at a time so you know what improved the result.
Portraits with a point of view
1. Door peephole portrait
Output: image.
Using the attached reference photograph, create one vertical image. Place the portrait inside a circular apartment peephole view, with curved hallway lines and a dark outer rim. Keep the face near the optical centre so the distortion belongs mainly to the surrounding architecture.
2. Train reflection double scene
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Seat the person beside a train window. Let reflected platform lights drift across the glass while their face remains softly visible behind it. Keep the carriage stable and animate the passing reflections rather than rotating the person. No added captions or audio.
3. Mirror introduction
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Frame the person from behind as they lean slightly into a bathroom mirror. Reveal their reflected face with that small movement. Match the reflected gesture and clothing precisely; keep the camera fixed and exclude extra reflections. No added captions or audio.
4. Through a camera viewfinder
Output: image.
Using the attached reference photograph, create one vertical image. Compose an over-the-shoulder photograph of a second camera framing this person. Use a simple unbranded optical viewfinder border without invented readouts. Keep the subject recognisable and let the foreground camera remain softly out of focus.
5. Shoulder glance street portrait
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Begin with the person facing away at a quiet crossing, then have them glance gently towards the lens without turning their body. Animate one background pedestrian in the distance; preserve facial identity and avoid busy crowd movement. No added captions or audio.
6. Wind through one sleeve
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Keep the person still against a plain wall while a light breeze moves one loose sleeve and a few hair strands. Their expression remains unchanged. Use soft side lighting so the fabric motion becomes the only visual event. No added captions or audio.
7. Window-seat shadow portrait
Output: image.
Using the attached reference photograph, create one vertical image. Place the seated person beside a large window whose geometric shadow falls across the chair and floor. Keep the face in clean indirect light. Build the composition around the diagonal shadow rather than placing stripes over their eyes.
8. Headphones quiet moment
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied portrait with visible headphones and animate a small relaxed head tilt followed by stillness. Keep the headphones attached naturally and their shape unchanged. Background activity stays blurred; do not invent singing or spoken dialogue. No added captions or audio.
9. Desk chair arrival
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Begin with the person just beside a desk chair, then show them settle into it and look towards an offscreen monitor. Keep hands relaxed and mostly below frame. Let the chair move once without changing its construction. No added captions or audio.
10. Portrait behind rain glass
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Place the person behind a window with several raindrops in the foreground. Focus on the glass first, then transfer focus slowly to their face. Keep their features stable and animate only a few drops travelling down the pane. No added captions or audio.
Graphic portrait treatments
11. Two-colour risograph portrait
Output: image.
Using the attached reference photograph, create one vertical image. Translate the portrait into a two-ink risograph print with coral and deep blue layers, slightly offset registration and visible paper texture. Preserve the facial proportions and hairstyle. Keep the background as one clean block with no generated lettering.
12. Charcoal study with one accent
Output: image.
Using the attached reference photograph, create one vertical image. Render the portrait as a careful charcoal drawing on warm paper, retaining the person's expression and recognisable features. Leave the clothing loosely sketched and add one restrained orange pencil mark beside the composition, away from the face.
13. Portrait contact sheet selection
Output: image.
Using the attached reference photograph, create one vertical image. Create a six-frame contact sheet using the same supplied portrait crop, with small differences in crop and exposure only. Keep pose and identity identical. Add blank frame margins without fabricated film metadata, then circle one frame in pencil.
14. Halftone portrait window
Output: image.
Using the attached reference photograph, create one vertical image. Place a monochrome halftone version of the portrait inside one large irregular editorial window. Keep the rest of the canvas cream with generous negative space. Preserve expression and hair shape; use dot density to describe shading rather than extra colours.
15. Cut-paper silhouette study
Output: image.
Using the attached reference photograph, create one vertical image. Reduce the portrait to overlapping flat paper shapes for hair, face and clothing, with clearly cut edges and subtle layer shadows. Keep the silhouette recognisable. Avoid reconstructing detailed facial anatomy; use the original pose as the composition anchor.
16. Magazine margin portrait
Output: image.
Using the attached reference photograph, create one vertical image. Set the portrait as a narrow vertical crop on the right of a cream editorial page. Place three thin horizontal rules and empty text areas on the left. Preserve the photograph without fabricating magazine names, quotations or publication credentials.
17. Blueprint-free contour portrait
Output: image.
Using the attached reference photograph, create one vertical image. Trace the person's silhouette and major clothing folds with delicate white contour lines over a darkened version of the original photograph. Leave the face photographic and readable. Add no measurement marks or technical claims; this is a graphic overlay.
18. Ink wash background separation
Output: image.
Using the attached reference photograph, create one vertical image. Keep the person photographic while translating only the original background into pale watercolour washes. Preserve the boundary around hair and glasses. Let pigment bloom near the outer edges, with no paint covering the subject's face or clothing details.
19. Portrait shadow theatre
Output: image.
Using the attached reference photograph, create one vertical image. Show the portrait as a clean photographic cutout beside an enlarged flat silhouette on a contrasting wall. Both must match the original pose. Use the second silhouette as a graphic compositional element, without suggesting a different person is present.
20. Duotone newspaper crop
Output: image.
Using the attached reference photograph, create one vertical image. Reproduce the portrait as a close newspaper-style duotone crop on softly folded paper. Maintain eye alignment and facial proportions. Leave the surrounding columns blank for later typesetting, and avoid adding a newspaper masthead or invented report about the person.
Products through practical optics
21. Split diopter product study
Output: image.
Using the attached reference photograph, create one vertical image. Compose the product sharply in the near right foreground and one neutral prop sharply in the distant left background, separated by a deliberately soft transition zone. Use a split-diopter photographic look without doubling the product or changing its front label.
22. Prism edge rainbow
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Keep the product stationary as a small prism moves just outside frame, sending a narrow rainbow along the background and tabletop. Prevent the coloured band from obscuring the label. Lock the camera and keep reflections physically consistent. No added captions or audio.
23. Magnifying glass detail
Output: image.
Using the attached reference photograph, create one vertical image. Place a handheld magnifying glass over one visible surface detail, enlarging only that area through the lens. Keep the surrounding product at normal scale. Use the supplied detail as reference and do not invent microscopic textures or printed information.
24. Frosted divider reveal
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Begin with a frosted glass divider obscuring the product. Slide the divider sideways until the untouched front view becomes clear. Keep the product and camera stationary, preserve the label, and show a believable edge on the moving glass. No added captions or audio.
25. Water-glass refraction border
Output: image.
Using the attached reference photograph, create one vertical image. Place a clear drinking glass in the extreme foreground, refracting a small part of the background beside the product. Keep the product's main face unobstructed. Use the optical distortion as an edge detail rather than bending the packaging itself.
26. Pinhole light study
Output: image.
Using the attached reference photograph, create one vertical image. Photograph the product in a softly darkened room with a tiny warm pool of light surrounding its base. Use gentle optical falloff and fine photographic grain. Keep the front label legible and avoid simulated camera-interface text or measurement markings.
27. Lens hood reveal
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Start with a dark circular foreground obstruction around the product, then move the obstruction away to reveal the complete rectangular composition. Treat it as a physical lens hood near the camera. Keep the product fixed and end on a clean hold. No added captions or audio.
28. Polarised reflection comparison
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Keep a glossy product still while the surface glare gradually reduces, revealing the existing label more clearly. Do not change the material or colour beneath it. Hold lighting direction and framing constant so the difference reads as reflection control. No added captions or audio.
29. Foreground bokeh passage
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Move a short strand of out-of-focus lights across the extreme foreground while the product stays tack sharp behind it. Let only two or three soft circles cross the edge of the frame. Preserve the label and avoid covering the entire object. No added captions or audio.
30. Mirror strip multiplication
Output: image.
Using the attached reference photograph, create one vertical image. Place two narrow mirror strips beside the product to create partial side reflections, while the original front view remains central. Keep reflected lettering appropriately mirrored. Do not generate unseen rear details; limit reflections to surfaces supported by the supplied photographs.
Small tabletop mechanisms
31. Sliding colour cards
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Keep the product stationary on a neutral table while two coloured cards slide into position behind it from opposite sides. Their edges meet off-centre to form a graphic backdrop. Use real paper thickness and keep all movement behind the object. No added captions or audio.
32. Bookend frame closure
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Place the product between two simple upright bookends that slide slightly inward and stop with generous clearance. Keep its scale and position fixed. Use the motion to frame the product without touching, compressing or altering any part of it. No added captions or audio.
33. Index-card staircase
Output: image.
Using the attached reference photograph, create one vertical image. Build a small staircase from thick blank index cards beside the product, with their clean edges stepping towards its visible feature. Keep the product resting on the table, not on unsupported paper. Leave every card blank for later annotation.
34. Accordion backdrop extension
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Keep the product in front of a folded paper backdrop as the accordion expands gently sideways. Maintain consistent fold widths and shadows. The product remains static and fully visible, with the paper movement creating depth only in the background. No added captions or audio.
35. Pegboard accessory organisation
Output: image.
Using the attached reference photograph, create one vertical image. Arrange the supplied product and only its photographed accessories on a pale pegboard with simple hooks. Show clear separation and believable support. Use the reference images for every accessory; leave uncertain spaces empty rather than inventing included components.
36. Balance beam composition
Output: image.
Using the attached reference photograph, create one vertical image. Place the product on the broad fixed end of a sculptural balance-beam display, with a neutral counterweight at the opposite side. Show a stable support underneath. Keep this an editorial set, not a demonstration of the product's actual weight.
37. Rolling spool backdrop
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Keep the product fixed while a small spool rolls across the distant tabletop, leaving one loose thread line behind it. Keep the thread away from the label and moving parts. Use restrained motion and let the spool settle before the clip ends. No added captions or audio.
38. Shadow-box insert
Output: image.
Using the attached reference photograph, create one vertical image. Place the front-facing product inside a shallow open shadow box whose inset edges create layered rectangular shadows. Keep the recess large enough for the supplied proportions. Use warm paperboard surfaces and avoid depicting the display as actual retail packaging.
39. Sliding shelf divider
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Move one empty shelf divider laterally to reveal the product already sitting behind it. Keep the camera square to the shelf and all lines straight. The reveal ends with the original front face unobstructed and no product rotation. No added captions or audio.
40. Elastic frame tension
Output: image.
Using the attached reference photograph, create one vertical image. Surround the product with thin coloured elastic cords stretched across a separate open frame. Keep every cord clear of the product silhouette. Show visible attachment points on the frame so the graphic lines feel like a constructed studio set.
Hands doing one clear thing
41. Bookmark placement
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied open-book photo and show one hand place a plain bookmark between the visible pages, leaving its top edge exposed. Keep printed content unchanged and avoid generating new readable passages. Hold overhead framing and end before the book closes. No added captions or audio.
42. Mug handle grip
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied mug photo and show one hand approach, place fingers naturally around the visible handle, then pause without lifting. Keep the handle shape and mug position fixed. Use a tight crop and avoid steam unless already present. No added captions or audio.
43. Notebook tab addition
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied notebook photo and show a hand attach one blank coloured page tab to the outer edge. Keep page count and binding unchanged. The tab should bend slightly under the fingertip, then remain in place after the hand leaves. No added captions or audio.
44. Fabric seam inspection
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied garment close-up and show two fingertips gently lift the edge of a visible seam, then release it. Preserve stitching and material grain. Keep the movement small enough to avoid exposing an invented lining or hidden construction. No added captions or audio.
45. Sunglasses fold
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use a reference showing the sunglasses and both temples. Animate one hand folding a single temple along its visible hinge while the frame stays supported. Keep lens shape and reflections stable; do not add a second fold or a dramatic camera move. No added captions or audio.
46. Pen clip attachment
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use supplied photographs of the pen and notebook. Show a hand slide the pen's visible clip over the notebook cover edge, then withdraw. Preserve the clip geometry and realistic scale. Stop if the references do not show how the clip attaches. No added captions or audio.
47. Envelope alignment
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied stationery flat lay and show one hand square a blank envelope against the edge of a notebook. Keep the camera overhead, paper surfaces matte and shadows soft. Do not open the envelope or invent personal names and addresses. No added captions or audio.
48. Jewellery clasp display
Output: image.
Using the attached reference photograph, create one vertical image. Use a close reference of the actual clasp and show fingertips holding it open for inspection. Keep the visible mechanism faithful to that reference. Focus on the clasp and fingertips, with the remaining jewellery softly receding; do not invent construction details.
49. Keyboard single-key action
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied keyboard photo and show one fingertip press a clearly visible key once, then release. Preserve the existing key legends and spacing. Keep the camera locked and do not animate an unrelated screen or imply that software completed a task. No added captions or audio.
50. Box insert placement
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied open box and insert photographs. Show one hand place the matching insert into the empty box with a slow straight movement. Preserve the actual fit and visible packaging design. Do not create additional accessories, printed claims or hidden compartments. No added captions or audio.
Food with restrained movement
51. Toast crumb trail
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied toast photo and arrange a short trail of loose crumbs leading towards its cut edge. Keep the topping and portion unchanged. Light from the side to reveal texture without making the food look wetter or more heavily dressed.
52. Fork twirl close-up
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied pasta image and show one fork gather a small visible portion with a controlled half-turn. Keep the plate stationary and sauce consistent. Avoid extra utensils, impossible noodle stretching or changing the quantity of food during the motion. No added captions or audio.
53. Tea-bag lift
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied cup photo with a visible tea bag. Lift the bag slowly by its string, allowing a few drops to return to the cup. Keep the liquid level plausible and label blank unless its design is supplied. No added captions or audio.
54. Citrus peel ribbon
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied citrus photo and place one loose peel ribbon beside the fruit, following a gentle curve. Match the visible rind texture and thickness. Keep the fruit intact unless a cut reference is provided; avoid implying an ingredient relationship to unrelated products.
55. Spoon surface indentation
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied yoghurt or soft dessert photo and show a spoon press gently into the existing surface, leaving one shallow indentation. Preserve the bowl and topping arrangement. Keep the movement local and avoid scooping out an invented interior filling. No added captions or audio.
56. Ice cube rotation in glass
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied drink photo with visible ice and animate one cube turning slowly near the surface as the liquid settles. Preserve the glass outline and fill level. Keep bubbles and garnish unchanged; avoid a splash or new ingredients. No added captions or audio.
57. Sandwich paper fold
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied sandwich photo and show a hand fold one edge of the wrapping paper down to reveal more of the existing filling. Keep the sandwich still and ingredients unchanged. End on the exposed edge without taking a bite. No added captions or audio.
58. Cookie cooling-rack geometry
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied cookie photo to arrange three matched cookies on a wire cooling rack, with diagonal rack shadows across the background. Preserve their original surface texture and approximate size. Do not invent a broken centre or additional filling.
59. Coffee crema line
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied espresso photo and animate a tiny settling movement along the existing crema boundary. Keep the cup and liquid level fixed. Use a close overhead frame with natural reflections, avoiding invented latte art or an apparent pouring action. No added captions or audio.
60. Berry fingertip selection
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied bowl-of-berries photo and show fingertips pick up one visible berry near the rim. Keep the remaining fruit in place and preserve natural contact shadows. Use a close crop, with no taste reaction, added syrup or exaggerated shine. No added captions or audio.
Fashion and accessory detail
61. Bag strap shadow arc
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied bag photo and arrange its visible strap into a loose arc on a matte surface, casting a clean shadow beside the bag. Preserve the hardware and stitching. Keep the bag closed and do not invent interior compartments.
62. Scarf thread movement
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied scarf close-up and animate a slight flutter at its fringe while the woven body stays still. Keep fibre texture and colour stable. Use raking soft light and a fixed camera so the movement communicates material rather than a scene change. No added captions or audio.
63. Watch cuff reveal
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied wristwatch and sleeve photographs to show the cuff move back slightly, revealing the watch face. Keep time markings and case shape unchanged. Avoid a rotating wrist or invented brand lettering; finish with the face clearly visible. No added captions or audio.
64. Shoe lace settling
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied shoe photo and animate the loose lace ends settling onto the tabletop after a tiny offscreen release. Preserve the knot, eyelets and shoe silhouette. Keep the shoe still and avoid presenting the movement as an automatic lacing feature. No added captions or audio.
65. Ring velvet impression
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied ring photo and place it in a shallow depression in velvet, with the fabric pile bending naturally around its base. Preserve the gemstone shape and setting. Show a front-supported angle rather than inventing the underside of the mount.
66. Jacket lapel adjustment
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied jacket portrait and show one hand lightly straighten the visible lapel, then leave frame. Preserve the person's pose, jacket seams and button count. Keep the motion local, with no wardrobe transformation or sudden lighting change. No added captions or audio.
67. Belt hardware crop
Output: image.
Using the attached reference photograph, create one horizontal image. Use the supplied belt close-up to create a horizontal composition with the buckle on one side and the strap extending across the frame. Preserve visible holes, finish and engraving. Leave clear negative space above without inventing measurements or durability claims.
68. Earring light glint
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied earring portrait and animate a tiny natural head movement that causes one restrained highlight on the existing metal edge. Preserve the ear, attachment and jewellery geometry. Avoid swinging through unseen angles or generating additional stones. No added captions or audio.
69. Folded knit rhythm
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied knitwear photograph to create a tidy folded arrangement with one sleeve crossing diagonally over the body. Preserve the stitch pattern, colour and visible label. Keep the folds plausible for the material, without inventing a second garment.
70. Zipper partial opening
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use a close reference showing a real zipper and its surrounding fabric. Move the slider a short distance along the visible teeth, revealing only the supplied lining reference. Keep the motion straight and stop before any unsupported interior becomes visible. No added captions or audio.
Workspaces and architecture
71. Monitor wake reflection
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied desk photo and let a blank offscreen monitor brighten gradually, adding a soft rectangular reflection to nearby surfaces. Keep every object fixed. Show no invented software interface, messages or analytics; the effect is entirely a change in illumination. No added captions or audio.
72. Office blind adjustment
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied office photo and tilt the visible blind slats slightly so the daylight pattern on the floor becomes softer. Preserve room geometry and furnishings. Keep the camera stationary and avoid moving the sun across an unrealistic distance. No added captions or audio.
73. Chair shadow composition
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied room photo and isolate one existing chair against a clean wall, with a long diagonal shadow defining the empty space. Preserve the chair's construction and proportions. Remove clutter without inventing an architectural feature or a new doorway.
74. Curtain breathing loop
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied room photograph and animate the lower edge of a sheer curtain drifting gently inward, then returning. Keep furniture and window frames fixed. Maintain the same daylight throughout and make the first and last positions suitable for a subtle loop. No added captions or audio.
75. Elevator threshold view
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied lobby photo and frame it from an open elevator doorway, using the dark doorway as a foreground border. Preserve the lobby's visible architecture. Do not invent company signage, building credentials or a reflection that reveals unseen camera equipment.
76. Desk cable organisation
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied desk photograph to gather visible cables into a tidy route along the desk edge with simple clips. Preserve each device and its actual connections where visible. Hide uncertain cable endpoints rather than inventing ports or claiming electrical modifications.
77. Door handle attention
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use a supplied close-up of a door handle and show a hand gently depress the handle without opening the door. Preserve its mechanism and contact with the frame. Keep the background still and avoid revealing a room not present in the reference. No added captions or audio.
78. Sunroom floor reflection
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied sunroom photo and animate faint reflected leaf movement across the floor, as if outdoor foliage moved in a breeze. Keep the plants inside and all furniture still. Preserve window geometry and avoid a sweeping camera move. No added captions or audio.
79. Stairwell graphic crop
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied staircase photo to create a tightly cropped architectural composition of its existing handrail, steps and shadows. Preserve the actual geometry and safe-looking connections. Emphasise diagonal rhythm without adding impossible steps, extra floors or decorative text.
80. Desk calendar page lift
Output: video.
Using the attached reference photograph, create a short vertical video in one continuous shot. Use the supplied desktop calendar photo and lift one blank rear page slightly in a light breeze before it settles. Preserve the front page and any supplied dates. Keep the binding coherent and avoid generating a new month or changing readable information. No added captions or audio.
Editorial layouts and useful assets
81. Five-frame storyboard sheet
Output: image.
Using the attached reference photograph, create one vertical image. Use the supplied photograph as the visual reference for a five-frame storyboard sheet, varying crop and composition while retaining the same scene. Leave caption strips blank. Keep it visibly an editorial planning asset rather than a claim that five real photos exist.
82. Subject-safe title canvas
Output: image.
Using the attached reference photograph, create one vertical image. Recompose the supplied photo into a vertical title-card background with the subject in the lower right and a clean upper-left reading area. Preserve identity and object details. Extend only simple background surfaces, leaving text out for accurate later typesetting.
83. Photo annotation margin
Output: image.
Using the attached reference photograph, create one vertical image. Keep the supplied photograph intact inside a wide cream margin with three fine leader lines ending beside visible details. Leave their labels blank. Avoid arrows into hidden components or invented performance claims; make this a reusable explanatory layout.
84. Colour study from the scene
Output: image.
Using the attached reference photograph, create one vertical image. Place the supplied photograph above five clean colour swatches sampled from its visible palette. Leave all hex labels blank for later verification. Keep the photograph's colours unchanged and make the swatches secondary to the image rather than a new product colourway.
85. Document-and-detail diptych
Output: image.
Using the attached reference photograph, create one vertical image. Create a two-panel composition from the supplied document photo: the complete page on one side and a crop of one existing visual detail on the other. Preserve readable content exactly where possible; leave uncertain text unchanged rather than recreating it.
86. Portrait lower-third background
Output: image.
Using the attached reference photograph, create one vertical image. Reframe the supplied portrait with clear space across the bottom quarter for later subtitles. Keep face and hands outside that zone. Use a subtle local darkening behind the empty subtitle area while preserving the person's skin tone and the surrounding photograph.
87. Resource card photo strip
Output: image.
Using the attached reference photograph, create one vertical image. Crop the supplied workspace image into three equal horizontal photo strips separated by cream gutters. Use distinct crops from the same photograph rather than inventing new objects. Leave a blank caption band under each strip for accurate labels added later.
88. Circular avatar proof sheet
Output: image.
Using the attached reference photograph, create one vertical image. Show the supplied portrait in three circular crops at different zoom levels on a plain background. Preserve the exact identity and expression in each. Include a tiny-size preview beside each crop so the reader can compare recognition without generated text.
89. Landscape-to-story adaptation
Output: image.
Using the attached reference photograph, create one vertical image. Adapt the supplied landscape photograph into a vertical story composition by extending simple sky or wall areas above and below. Keep the original scene centred and undistorted. Do not add people, signage or landmarks; leave the added space available for later copy.
90. Image-led chapter divider
Output: image.
Using the attached reference photograph, create one horizontal image. Create a wide chapter-divider layout from the supplied photo, with a narrow image section on the left and a generous cream area on the right. Add one thin orange rule only. Preserve source details and leave all chapter names and numbers blank.
Transitions between supplied photos
91. Match the hand position
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use both supplied photos as start and end references. Match the visible hand position between them, then transition as the hand briefly passes close to the lens. Preserve each source scene's identity and keep the covered interval short, without adding extra fingers. No added captions or audio.
92. Doorframe matched cut
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use the two supplied room photos and create a transition as a dark doorframe fills the view between them. Match camera height and direction on both sides. Keep architecture faithful within each shot and avoid visibly morphing one room into the other. No added captions or audio.
93. Notebook cover match
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use the two supplied desk photos, each containing a notebook. Move into the first notebook's plain cover until it fills frame, then pull back from the second. Preserve both desk arrangements and use the cover surface to conceal the transition. No added captions or audio.
94. Light-to-light match
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use the two supplied photos with a visible lamp in each. Approach the first lamp's bright shade, transition during the brief bright frame, then reveal the second scene. Avoid flashing repeatedly and preserve the lamps' separate designs rather than transforming their hardware. No added captions or audio.
95. Vertical seam handoff
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use the two supplied product photos and divide the frame with a moving vertical seam. Keep each photograph geometrically intact on its own side as the seam travels. Match product height, preserve labels and use no invented intermediate product construction. No added captions or audio.
96. Shape-matched window
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use two supplied photographs containing a similar rectangular window or frame. Move towards the first rectangle, then emerge from the second at matching scale. Keep the transition brief and the two environments distinct, without creating an impossible continuous building. No added captions or audio.
97. Page-edge turn transition
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use two supplied photographs as images printed on separate facing pages. Turn one page to reveal the second image underneath, preserving each photograph's contents. Show believable paper thickness and a single page movement, with no generated magazine text or brand masthead. No added captions or audio.
98. Colour patch handoff
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use two supplied photos sharing one large colour area. Move into that area in the first photo, then reveal the second through a matching colour field. Preserve subject identity within each image and avoid a visible face or product morph. No added captions or audio.
99. Triptych sequential replacement
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use three supplied photos in equal vertical panels. Replace the first panel, then the second and third, using a simple downward slide inside each boundary. Keep every photo intact, gutters fixed and motion slow enough to read as deliberate editorial sequencing. No added captions or audio.
100. Printed-photo shuffle
Output: video.
Using the attached reference photographs, create a short vertical transition video. Use three supplied photos as separate small prints on a table. Show a hand slide the top print aside to reveal the next, then stop. Preserve each image without rewriting details, keep contact shadows realistic and avoid inventing text on the print borders. No added captions or audio.
Make the next version more specific
If the result looks generic, change one concrete thing: the camera position, the motion, the light direction or the background. If the subject changes, shorten the movement and strengthen the source reference instead of adding more adjectives.
Before using a clip, inspect its beginning, middle and end. Look for drifting facial features, label changes, extra fingers, moving contact points and objects passing through each other. For stills, inspect at full size and at the size people will see on their phone. Add important text and logos in your design editor when exact reproduction matters.
Keep the source, successful prompt and selected settings together so you can reproduce the direction later. Start with the effect that supports your message, then make one version you can actually use. More practical resources are in all free guides.