Turn one adult portrait into a glowing carved pumpkin head video, or switch to a still photo. Keep your outfit and setting as references, choose a carving, and review credits before generating.
Third-party silent transformation reference from Vidney. Not a CreateForge result; the prompts and generated sound have not been tested with real generation.
Prompts to try
Classic triangle eyes
Video: a gradual head change and gentle tilt. Follow the selected duration and framing.
Image 1 defines one adult, their body, skin tone, clothes, hands, pose and surroundings. Keep the same subject and visible outfit, without slimming, changing age or adding people. Keep the original background as the scene reference; adapt framing only to the selected aspect ratio. Video 1 guides only the head transformation and gentle tilt, never its actor, clothing, room or framing. In one continuous shot using the selected duration, first show the person's ordinary face looking at the camera, then gradually turn only the head into a three-dimensional carved pumpkin. Keep a single head seated naturally on the shoulders with a short curved stem and warm light flickering inside the carved openings. After the change, slowly tilt the pumpkin to one side. No human face or skin inside the final pumpkin. Keep the hands natural and the body still. No floating pumpkin, duplicate head, victim, severing, blood, explosion, text or logos. If audio is enabled, use only newly generated soft wind and fabric sounds, with no music, lyrics or speech. Use an orange ribbed pumpkin with triangle eyes, a small triangle nose and a carved zigzag grin.
Friendly smile
Video: a gradual head change and gentle tilt. Follow the selected duration and framing.
Image 1 defines one adult, their body, skin tone, clothes, hands, pose and surroundings. Keep the same subject and visible outfit, without slimming, changing age or adding people. Keep the original background as the scene reference; adapt framing only to the selected aspect ratio. Video 1 guides only the head transformation and gentle tilt, never its actor, clothing, room or framing. In one continuous shot using the selected duration, first show the person's ordinary face looking at the camera, then gradually turn only the head into a three-dimensional carved pumpkin. Keep a single head seated naturally on the shoulders with a short curved stem and warm light flickering inside the carved openings. After the change, slowly tilt the pumpkin to one side. No human face or skin inside the final pumpkin. Keep the hands natural and the body still. No floating pumpkin, duplicate head, victim, severing, blood, explosion, text or logos. If audio is enabled, use only newly generated soft wind and fabric sounds, with no music, lyrics or speech. Use an orange ribbed pumpkin with rounded carved eyes and a friendly curved smile, with no frightening teeth.
White pumpkin
Video: a gradual head change and gentle tilt. Follow the selected duration and framing.
Image 1 defines one adult, their body, skin tone, clothes, hands, pose and surroundings. Keep the same subject and visible outfit, without slimming, changing age or adding people. Keep the original background as the scene reference; adapt framing only to the selected aspect ratio. Video 1 guides only the head transformation and gentle tilt, never its actor, clothing, room or framing. In one continuous shot using the selected duration, first show the person's ordinary face looking at the camera, then gradually turn only the head into a three-dimensional carved pumpkin. Keep a single head seated naturally on the shoulders with a short curved stem and warm light flickering inside the carved openings. After the change, slowly tilt the pumpkin to one side. No human face or skin inside the final pumpkin. Keep the hands natural and the body still. No floating pumpkin, duplicate head, victim, severing, blood, explosion, text or logos. If audio is enabled, use only newly generated soft wind and fabric sounds, with no music, lyrics or speech. Use an ivory-white ribbed pumpkin with triangle eyes and a simple carved smile; retain warm amber light inside.
01
Choose one clear portrait
Use your own photo or an adult's photo with permission. Show the outfit, hands and space above the head.
02
Choose video or photo
Video starts with your face, transforms the head and adds a gentle tilt. Photo makes the finished pumpkin look in a single image.
03
Review the quote and result
Check credits before submitting. Inspect the pumpkin, shoulders, hands and background before opening the full result or downloading.
What does AI pumpkin head make?
A carved jack-o'-lantern takes the place of your head, with a stem, ribbed surface and a warm glow inside its eyes and mouth. The video begins with the person, shows the change and ends with a small tilt. The photo mode makes the finished look directly. This page focuses on the pumpkin head transformation, rather than a foreground pumpkin hiding a dancer's head.
The intended edit keeps the body, clothing, hands, pose and original setting as references. AI redraws the scene, so fabrics, lighting and small background details can still change. Once the transformation is complete, the face is covered; this is not a face-preservation portrait effect.
Prepare your face, shoulders and outfit
Choose one adult with a clear face and even light, ideally from the waist or knees up. Leave room above the head for a pumpkin larger than the original head. A hand across the face, a tight crop or an oversized hat can make the edges harder to follow. A full-body photo works too, provided the head remains readable.
Use a supported frame that suits your source. The default output is vertical, but a wide source may need reframing or model-invented surroundings. The included horizontal reference guides the action only; it does not require your room, clothes or framing to match its example.
Three pumpkin prompts, two separate modes
Classic triangle eyes uses an orange pumpkin and zigzag grin. Friendly smile uses rounded eyes and a gentle carved mouth. White pumpkin uses an ivory shell with warm amber light. Apply or copy the complete prompt without generating, or edit it in Advanced prompt.
Video uses Seedance 2.0; photo uses GPT Image 2. Each mode has its own quote and explicit Generate action. Switching modes keeps a compatible photo but changes the model settings and clears the old quote. It never buys an image first and automatically charges for a video afterwards.
Motion reference, silence and credits
The included reference is about 5.04 seconds long; the default output is 5 seconds. Those are separate durations. The reference is uploaded for your account and its measured duration participates in the server quote. It is not a free input or a fixed price copied from another site.
The default video is silent. Add music on your publishing platform, or enable generated audio in Settings to request original wind and fabric ambience. No reference song, speech or audio file is submitted. The sample is labelled as a third-party reference, not a CreateForge output or proof that these prompts have passed creative testing.
Check the pumpkin before sharing
Look for a pumpkin resting naturally on the shoulders, one head, no face showing through the carved holes and hands separate from the rim. Floating pumpkins, duplicated heads, changing clothes or a missing before-and-after transition are possible model errors. Review the complete video or full-size photo rather than only the small preview.
A completed job does not guarantee an exact carving, identical background or precise transformation timing. The tool does not automatically repair, regenerate or charge again. If you decide to make another attempt, first simplify the prompt or choose a clearer source photo and review the new quote.
Credits and saved results
Sign in to upload references and review the current credit quote. Editing a prompt or changing a photo or setting clears the old quote. Generation starts only when you submit; opening this page or choosing a prompt does not spend generation credits.
Review the finished media before sharing. Successful results appear in Library with actual credit use. Result file URLs are public; browsing Library and task history still requires an account. References use temporary uploads, so a page refresh requires selecting photos again.
Common questions
It appears at the start of the video, then the pumpkin covers it. The finished photo also covers the face. Clothes and body help carry the original look, but exact preservation is not guaranteed.
This version is designed for one adult in one photo. It does not merge separate portraits or offer a pet template.
No. Each mode requires a separate click and quote. Choosing a template or switching modes does not generate anything.
Yes. Video mode includes a required motion reference whose measured duration is used in the server quote. Photo mode submits only your photo.
No. Audio is off by default. Optional generated audio asks for original ambience, not a copied song or voice.
Yes. The server quotes the selected model, settings and references before submission. This page does not promise a free generation or copy another website's credit prices.
References guide appearance, but likeness, hands and motion can vary. Use a clear photo and inspect the result. A completed task is not a guarantee that every creative detail is correct.