# AI Genjutsu

> AI Genjutsu (Genjutsu AI) is a browser-based AI video workspace at https://aigenjutsu.app. Recreate a reference scene with your own people, pets, or products using motion transfer, object swap, and ready-made presets.

Useful for creators who want a recognizable performance, pet dance, character recast, vehicle stunt, product replacement, or outfit transition with their own subjects.

- Motion transfer: use a reference performance to guide movement and character photos to guide appearance. Choose a preset or upload your own clip with a compatible workflow.
- Object swap: use supported video-edit workflows to replace a vehicle, prop, product, or outfit while following the reference scene. Choose a preset whose labeled image roles match what you want to change.
- Ready-made trends: the tools below supply the performance clip, prepared prompt, and photo roles. Add the requested images without searching for a source video or writing a prompt. Templates with reference audio restore that soundtrack after visual generation; silent references remain silent.
- Preview first: selecting a preset loads its setup; generation starts after submission. Check the credit estimate, then review the finished video in My creations before downloading. Likeness, hands, object edges, reflections, and fast movement can vary.
- Generation requires sign-in and sufficient credits. Model and output settings affect the cost; consult the selected tool and current Pricing page for estimates and download benefits.

AI Genjutsu is an independent service, unaffiliated with Higgsfield. Community showcases illustrate reference ideas and do not imply they were generated here. Use images, video, and audio you have permission to use.

## Start creating

- [Genjutsu AI workspace](https://aigenjutsu.app/): Start with Motion Transfer or Object Swap, browse presets, or bring your own reference clip and supported subject images.

## Photo-to-video trends

- [Tesla Roadster Meme](https://aigenjutsu.app/tesla-roadster-meme): A fictional police-chase story with your chosen driver. The police, vehicle and chase remain fixed. This is a full-character replacement, with the mask sequence following the prepared action.
- [Unforgettable Car AI](https://aigenjutsu.app/unforgettable-car-ai): A card-led reveal of a chosen person and vehicle. Required person and car roles stay distinct. Extra images customize only the supported background and card slots.
- [Ceiling Fan AI](https://aigenjutsu.app/ceiling-fan-ai): A four-look transformation timed to a passing fan blade. The fan and overhead camera provide the transition. Four inputs represent successive looks, not four people sharing the room.
- [Day by Day Dance AI](https://aigenjutsu.app/day-by-day-dance-ai): A dance that builds from one performer to a trio. The entrance order is fixed. A group photograph cannot replace the three independently assigned reference slots.
- [Zombie AI](https://aigenjutsu.app/zombie-ai): A two-character dramatic scene linked to warmer memories. The story beats, infection styling and original props follow the reference. There is no control for writing a different ending.
- [Boxing AI](https://aigenjutsu.app/boxing-ai): A living-room shadowboxing performance. The room and routine stay prepared. This is shadowboxing with no added opponent or editable fight choreography.
- [Ozil Goal AI](https://aigenjutsu.app/ozil-goal-ai): A football attack starring your chosen scorer. The other players, ball and pitch remain part of the reference. This page does not change the match or add a new goal sequence.
- [Bad Times AI](https://aigenjutsu.app/bad-times-ai): An expressive kitchen reaction from a pet photo. The performance follows existing mouth movements and sound. It does not turn typed dialogue into new pet speech.
- [Otsukare Summer Dog AI](https://aigenjutsu.app/otsukare-summer-dog-ai) / [Daijoubu Dog Dance](https://aigenjutsu.app/daijoubu-dog-dance): A cheerful pet dance in a pastel setting. The scene, cuts and paw choreography are fixed. You do not upload separate dancers or select another song.
- [Cheba Man AI](https://aigenjutsu.app/cheba-man-ai): A crowd-framed dance led by your character. The onlookers and venue remain in the reference. This page provides one lead role.
- [Money Spread AI](https://aigenjutsu.app/money-spread-ai): A short cash-display performance. The money display is part of the template. The output should not be presented as proof of real earnings or financial results.
- [Roadster AI](https://aigenjutsu.app/roadster-ai): A ramp stunt performed by your selected vehicle or object. Unusual objects may deform as the model adapts them to vehicle motion. A person photo is not a driver slot here.
- [Flying Tesla Roadster AI](https://aigenjutsu.app/day-1-of-owning-a-roadster-ai): A flying-car joke ending with spilled coffee. The car is retained. This page replaces the driver, while the separate Roadster AI page replaces the vehicle or object.
- [Phone Stuck to Train AI](https://aigenjutsu.app/phone-stuck-to-train-ai): A platform performance seen from a departing train viewpoint. The train, platform and viewpoint are fixed. You do not need to film near a railway.
- [Bull Backflip AI](https://aigenjutsu.app/bull-backflip-ai): A fictional rodeo flip starring your character. The bull stays in the sequence. Upload a rider, and treat the result as a fictional visual effect.
- [Alors on Danse AI](https://aigenjutsu.app/alors-on-danse-ai): A studio duet with two clearly different personalities. The keyboard and laptop belong to the scene. Your second upload controls the partner, not a new instrument or a background.
- [Thriller AI](https://aigenjutsu.app/thriller-ai): A night-time group dance led by your chosen character. The template contains a short excerpt. It does not recreate an entire music video or let you arrange a different dance.
- [Bottle Kick AI](https://aigenjutsu.app/bottle-kick-ai): A comic trick attempt that ends in scattered bottles. The ending knocks the bottles apart. This is not a bottle-cap-opening template, and the stand cannot be replaced with an uploaded object.
- [Car Exit Salute AI](https://aigenjutsu.app/car-exit-salute-ai): A quick arrival and greeting beside a sports car. The red car and street remain in the reference. The vehicle is not an upload slot on this page.
- [Luxury Car Arrival AI](https://aigenjutsu.app/luxury-car-arrival-ai): A styled hotel entrance with a dressing moment. The added outer layer is part of the action. A result is not expected to hold one unchanged outfit throughout the clip.
- [Enzo Celebration AI](https://aigenjutsu.app/enzo-celebration-ai): A football celebration centered on your character. This sequence is the celebration. It does not add a scoring run or change the team around the lead.
- [Popstar AI](https://aigenjutsu.app/popstar-ai): A mansion performance with a headline entrance. The guard and bottle remain scene elements. There is no second cast slot or editable dialogue on this page.
- [Kung Fu Kick AI](https://aigenjutsu.app/kung-fu-kick-ai): A martial-arts action clip built around a moving kicker. The surrounding pad holders are retained. The video is a visual performance, not instruction for attempting the move.
- [Talk Dirty Dance AI](https://aigenjutsu.app/talk-dirty-dance-ai): A concert appearance with you leading the formation. The surrounding dancers and fire choreography are fixed; this page has no slots for recasting the whole troupe.
- [Bottle Monologue AI](https://aigenjutsu.app/bottle-monologue-ai): A roadside delivery with the original spoken performance. The soundtrack and mouth movements follow the source. Writing new speech or cloning a voice is outside this template.
- [Jean Phil Dance AI](https://aigenjutsu.app/jean-phil-dance-ai): A playful solo in a small white room. Several clips share this trend name. This page recreates the particular room routine shown in the preview.
- [Bby Wow AI](https://aigenjutsu.app/bby-wow-ai): A party entrance with a playful camera-facing beat. Guests are retained and the party choreography is fixed. This page does not accept separate photos for the crowd.
- [Street Walk AI](https://aigenjutsu.app/street-walk-ai): A short moving portrait on a city street. The street location and camera movement come from the template; a second image cannot relocate the scene.
- [Cinematic Suit AI](https://aigenjutsu.app/cinematic-suit-ai): A recurring character across a compact cinematic montage. The input chooses the appearance, while props and edits are fixed. Longer external examples do not set this template's length.
- [Noir Suit AI](https://aigenjutsu.app/noir-suit-ai): A monochrome character montage with a formal mood. Upload a suited look if you want a suit. The original prop gestures and shot order remain part of the clip.
- [Elevator Dance AI](https://aigenjutsu.app/elevator-dance-ai): A multi-shot performance inside a mirrored elevator. The performance is an edited sequence, not one uninterrupted full-body dance. Mirrors do not create additional input roles.
- [Noir Edit AI](https://aigenjutsu.app/noir-edit-ai): A black-and-white montage following one character. The sequence and supporting people are fixed. A colorful external example does not mean this page offers a color-grade selector.
- [Peaky Blinders AI](https://aigenjutsu.app/peaky-blinders-ai): An unhurried estate walk against a dramatic explosion. The estate, background people and explosion timing stay prepared. The page has no official association with the television series.
- [Plug Walk AI](https://aigenjutsu.app/plug-walk-ai): A roadside exchange that becomes a relaxed dance. The hand inside the car is retained. You do not upload a second person to perform the handoff.
- [Dragon Encounter AI](https://aigenjutsu.app/dragon-encounter-ai): A fantasy meeting between your character and a towering dragon. The dragon is fixed. An uploaded creature image will not provide a separate dragon-replacement role.
- [Blueprint Reveal AI](https://aigenjutsu.app/blueprint-reveal-ai): An architectural character montage around plans and miniature buildings. Upload a person, not a blueprint or CAD drawing. This does not convert a real plan into an accurate building.
- [A2O LITE NEON AI Dance](https://aigenjutsu.app/a2o-lite-neon-ai-dance): A two-person studio dance from separate photos. The white studio and camera choreography are prepared. Matching clothes are optional, and the two uploads do not merge into one character.
- [Smalltown Boy AI](https://aigenjutsu.app/smalltown-boy-ai): A paired performance with two distinct characters. Both reference slots are required. Mixing two subjects in one image makes the intended assignments less clear.
- [Bulin 47 AI](https://aigenjutsu.app/bulin-47-ai): A lively lead performance in an indoor gathering. This page follows the particular performance in the preview. It does not create a new song, venue or crowd from your image.
- [Hotel Lobby AI](https://aigenjutsu.app/hotel-lobby-ai) / [AI Rap Duo](https://aigenjutsu.app/ai-rap-duo) / [Migos AI](https://aigenjutsu.app/migos-ai): A two-performer microphone scene in an orange studio. The microphone, orange studio and audio remain prepared. Additional images do not add a third lead.
- [Storm Ii AI](https://aigenjutsu.app/storm-ii-ai): A synchronized formation with one lead and a repeated supporting look. The second reference guides the group collectively. It does not assign a different unique person to every background position.
- [Raindance AI](https://aigenjutsu.app/raindance-ai): A seaside duet with two independently chosen performers. The coastal location and camera sequence are fixed. Your images select the performers rather than a new background.
- [gas station dance AI](https://aigenjutsu.app/gas-station-dance-ai): A three-role dance with a moving selfie viewpoint. The selfie operator returns at the end and remains the same role. The car does not have a separate replacement slot.
- [Put A Ring On It Dog](https://aigenjutsu.app/put-a-ring-on-it-dog): A pet-led studio routine backed by two dancers. The two human dancers stay in the reference. One pet photo supplies the sole replacement role.
- [car jump AI](https://aigenjutsu.app/car-jump-ai): A fictional seated-person and vehicle stunt. Both inputs are required; the reference supplies the chair, stunt path and setting.
- [Burning Bridges AI](https://aigenjutsu.app/burning-bridges-ai): A crowded indoor performance with a chosen lead. A lead image is required. Extra references belong to supporting appearances and do not create a new room or sequence.
- [Baile Charme AI](https://aigenjutsu.app/baile-charme-ai): A relaxed dance at the center of a night crowd. This is the specific night-party performance in the preview; the name can refer to other routines elsewhere.
- [Wipe Me Down AI](https://aigenjutsu.app/wipe-me-down-ai): A two-character dance relay in a kitchen. The roles follow the reference handover; changing the input order changes who takes each turn.
- [Window Peeper AI](https://aigenjutsu.app/window-peeper-ai): A tiny comedy about an interrupted meal. The food, window and curtain action remain prepared. The tool does not animate arbitrary meals or generate a new story from text.
- [Venom Car AI](https://aigenjutsu.app/venom-car-ai): A liquid-black reveal of your chosen vehicle. This input is a vehicle photo. It does not cast a person, create a technical assembly simulation or guarantee exact badge lettering.
- [Helicopter Cash AI](https://aigenjutsu.app/helicopter-cash-ai): A fictional cash-toss scene above the city. Only the foreground passenger is replaced. The rear passenger, helicopter and banknotes are retained.
- [Closed On Sunday AI](https://aigenjutsu.app/closed-on-sunday-ai): A quiet central character surrounded by an animated cat choir. The center stays quiet as directed by the reference. This is not a custom singing or voice-generation tool.
- [Te He Perdido Entre La Gente](https://aigenjutsu.app/te-he-perdido-entre-la-gente): A two-character recognition scene across a crowded club. The club crowd, blue lighting and camera reveal remain prepared. The two uploads do not replace everyone on the floor.
- [Connor Price Slide AI](https://aigenjutsu.app/connor-price-slide-ai): A street duet with a lead performer and a boombox partner. The boombox is part of the reference, and the soundtrack is fixed. This page does not generate a new song.
- [Gangnam Style AI](https://aigenjutsu.app/gangnam-style-ai): A three-person night-street selfie performance. This version uses the selfie routine in the preview. It does not promise a different dance simply because it shares the song name.
- [Migos Carpool AI](https://aigenjutsu.app/migos-carpool-ai): A four-person singalong inside a car. The car interior, seat belts and audio are fixed. Four seat assignments are required; this is separate from the two-person Hotel Lobby scene.
- [Hallway Dance AI](https://aigenjutsu.app/hallway-dance-ai): A four-person crew routine in a corridor. Opening clothing descriptions identify source roles. Your uploads define the new clothes even when positions change.
- [Trio Shuffle AI](https://aigenjutsu.app/trio-shuffle-ai): A three-character shuffle on a reflective night plaza. The assignments follow characters throughout the dance, not whichever person happens to occupy a screen position later.
- [Crosswalk Dance AI](https://aigenjutsu.app/crosswalk-dance-ai): A paired street-crossing performance. The supplied reference has no audio. This template creates a silent clip rather than adding a new soundtrack.
- [Crazy Frog AI](https://aigenjutsu.app/crazy-frog-ai): An impossible two-character street chase. The visible red vehicle remains in the scene. The first rider's invisible ride is part of the joke, not a missing car upload.
- [Epic Split AI](https://aigenjutsu.app/epic-split-ai): A cinematic split between two moving trucks. The trucks are fixed scene elements. This page creates an AI effect and is not guidance for attempting the stunt.
- [ATM Heist AI](https://aigenjutsu.app/atm-heist-ai): A fictional airborne action scene around a dragged atm. The ATM, chain and red car are retained. The template is a staged fantasy and supplies no real-world instructions.
- [Hug In Heaven](https://aigenjutsu.app/hug-in-heaven): An imagined reunion between two people. The scene is AI-generated symbolism. It does not reconstruct a real meeting, and the two photos should stay assigned to their original roles.
- [Street Duo AI](https://aigenjutsu.app/street-duo-ai): A playful two-person street routine. The original neon and skeleton outfits identify roles only. The uploaded clothes should guide the replacement appearances.
- [White Chicks AI](https://aigenjutsu.app/white-chicks-ai): A two-person comedy singalong in a convertible. The dog, convertible and source sound are retained. This is an independent template with no film affiliation.
- [Balloon Skydive AI](https://aigenjutsu.app/balloon-skydive-ai): An adventure sequence from balloon exit to open parachute. The flight path and parachute are fixed. A still photo does not provide a new route or a longer jump.
- [Archery AI](https://aigenjutsu.app/archery-ai): An archery-themed character montage. The montage does not depict a complete sporting demonstration or guarantee an arrow hitting a target.
- [Mountain Jump AI](https://aigenjutsu.app/mountain-jump-ai): A sudden move from a plaza interaction to a mountain view. The foreground partner stays unchanged. The title does not mean the output contains a cliff leap or a configurable destination.
- [Fridge POV AI](https://aigenjutsu.app/fridge-pov-ai): A snack-time joke seen from inside a refrigerator. The dessert and door sequence are fixed. It cannot substitute an uploaded food product or write a new eating routine.
- [Champagne Shoot AI](https://aigenjutsu.app/champagne-shoot-ai): A fictional rooftop celebration on a moving suv. The bottle throw and prop-pistol moment are part of the reference. This page does not remove individual beats or provide real stunt instruction.
- [City Giant AI](https://aigenjutsu.app/city-giant-ai): A city-scale visual effect starring your character. Buildings, traffic and viewpoints follow the reference. The upload does not select a real city or make a precise architectural simulation.
- [Upside Down AI](https://aigenjutsu.app/upside-down-ai): A hand-led transition into a second appearance. The reference has no soundtrack. Pink lyric text is part of the footage, and the second photo is another look rather than a different partner.
- [Car Assembly AI](https://aigenjutsu.app/car-assembly-ai): A stylized vehicle showcase in a desert setting. The effect is visual assembly imagery. It cannot generate accurate manufacturing steps or infer a mechanically valid build.

## Pricing

- [Pricing](https://aigenjutsu.app/pricing): Current plans, credit packs, and download benefits. Check the tool's estimate for the cost of a particular generation.

## Optional

- [AI Video Maker](https://aigenjutsu.app/video): Create a new clip from a prompt or supported image inputs.
- [Image to Video](https://aigenjutsu.app/video/image-to-video): Start with a still image and describe the subject or camera movement you want.
- [AI Image Maker](https://aigenjutsu.app/image): Create images from prompts or edit with supported reference images.
- [Terms](https://aigenjutsu.app/terms): Service terms and rights responsibilities.
- [Privacy](https://aigenjutsu.app/privacy): How uploaded media and account data are handled.
- [Refund policy](https://aigenjutsu.app/refund): Refund conditions and request process.