Left photo + right photo + motion video
Pair a character image with a performance reference and explore the orange-studio effect. Build a Hotel Lobby AI motion shot with clear inputs, inspect the result and choose how to finish the sequence.
Create my motion shotPrepare a duo image from your two photos, then add a motion reference. Sign in to upload and review the model and credit quote before generating.
Two performers, a suspended microphone and a simple orange backdrop.
Watch the collected performances to inspect framing, subject separation and changes of pose. Use them to understand the visual format; generated motion and character continuity still depend on the selected model and input media.
Two performers, a suspended microphone and a simple orange backdrop.
Contrasting outfits and an energetic performance in the same studio setup.
A playful pairing with a clear gap between the two subjects.
Costume silhouettes make the left and right performers easy to distinguish.
Illustrated characters bring a different visual treatment to the duo format.
An intergenerational pairing uses matching staging and contrasting identities.
A character and an animal sidekick share a readable performance space.
A studio performance with a suspended microphone and a warm orange backdrop.
Compare subject boundaries, crop and occlusion in the still frames before evaluating your own image and motion inputs.
Two performers, a suspended microphone and a simple orange backdrop.
Illustrated characters bring a different visual treatment to the duo format.
Contrasting outfits and an energetic performance in the same studio setup.
A playful pairing with a clear gap between the two subjects.
Prepare the visual input from a left photo and a right photo, combined into one duo image. Match the composition to your uploaded motion reference and assess the result for occlusion, framing and continuity. The selected model receives one image and one motion clip; two independent identity replacements are not part of this input contract.
Upload an MP4 or MOV, 3–30 seconds and no larger than 32 MB. Use a single readable performer with a stable crop. Choose a clip where the hands and face remain visible during the action you want the character to follow.
Use the left and right upload areas to select clear, similarly framed images. Prepare their combined reference and review subject boundaries. Select an available model and describe the studio and lighting without asking the prompt to override the reference action.
Read the displayed credit estimate before submitting. Review the full result for face, clothing and limb continuity, especially at changes of direction. Find the completed task in history and download the output for your finishing workflow.
Reduce conflicts between what the image shows and what the reference asks the character to do.
A waist-up reference needs useful information about the shoulders, torso and hands. A tightly cropped headshot may leave the model guessing. Choose a character image that exposes the same areas without unnecessary obstructions or additional people.
Rapid cuts, strong camera motion and performers crossing in front of one another make the movement harder to assess. Use a cleaner reference when possible. A short coherent action is often a better starting point than a complicated routine.
The location of a defect can help you decide which input needs attention.
If the face changes during a strong turn, inspect how much facial information the image provides and whether the reference hides it. A clearer image or a gentler motion reference may be more useful than repeating the same request.
Check whether arms cross the torso or small accessories disappear against the background. Use a simpler silhouette and an unobstructed gesture. There is no promise that exact garment details or every hand position survive motion transfer.
Generation and editing answer different questions about the final performance.
Generate each character with a shared studio brief, similar framing and contrasting clothing. Alternate the clips in an editor. This approach does not claim independent left and right replacement inside one task, and each output still needs continuity review.
Inspect the actual output dimensions and sound before choosing a publishing format. The original song, vocals and synchronized lip movement are not guaranteed. Use an authorized soundtrack and keep important gestures visible in the final crop.
Suggested prompts describe the intended look. They are not generated examples or guarantees of the result; motion still comes from the reference video.
Two original performers side by side in a minimal orange studio, clearly visible faces and shoulders, contrasting jackets, a suspended microphone between them, soft frontal lighting and a steady camera, movement guided by the uploaded performance reference.
One original stylized character with a clear silhouette and contrasting outfit in an orange recording booth, soft even lighting, uncluttered background, medium framing and expressive controlled motion following the uploaded reference.
Collected references illustrate the Hotel Lobby performance idea. Prepare your own permitted images and motion reference, then review the actual model and credit quote. Choosing a reference changes the preview; it does not import a preset or promise artist endorsement.
Use the left and right photo areas to prepare one side-by-side duo image, then add a motion reference. Both photos contribute to the composition. They are not sent as separate identity replacement controls, and the result depends on the selected model.
No unlimited or guaranteed free rendering is offered. Generation requires an available model and sufficient credits for the displayed quote. Check your balance and current pricing; promotional credits may not cover the chosen clip and settings.
For Higgsfield AI Hotel Lobby searches, compare required media, performer controls and output behavior. This page takes a motion clip and one character image. It does not claim identical templates, soundtrack treatment or generation cost.
You can prepare a character image in an image editor, then upload it in a supported format. Use a clear image you have permission to process; project files and editor settings are not imported.
A writing assistant can help organize a shot brief. Video rendering uses the selected model and its current credit quote; written planning does not provide a video preset or a separate generation integration.
A model must be available and have a supported quote for the reference duration and selected settings. Upload the reference and review the available controls. Do not treat an unavailable quote as a free generation or assume another model's price applies.
Exact soundtrack preservation and lip synchronization are not promised. Inspect the result and use music you are authorized to publish when finishing the edit. The reference and character image also need appropriate permission for processing.
It prepares one side-by-side image from both photos. That image is paired with your motion video. The left and right slots set the composition and do not create independent identity replacement inputs.
Match the character and motion inputs first. Review the moments where identity or anatomy changes, then decide which shot belongs in the final sequence.
Create my motion shot