Help & FAQ

MorphCraft is an AI platform for image and video transformations. You can swap faces, upscale images, remove backgrounds, apply artistic styles, animate photos, and generate images from text — all in one place.

Traditional editors require manual selection, masking, and layer manipulation. MorphCraft is AI-driven — you upload an image, choose a transformation, and the AI handles the rest. No design skills or software installation required.

No. MorphCraft runs entirely in the cloud. Just create an account and start transforming your images and videos in your browser.

Images: JPG, PNG, WebP (up to 20 MB per file). Videos: MP4, WebM (up to 100 MB per file). For best results use high-quality, well-lit source images.

Not yet. We are tracking demand for a developer API. If this is important to you, let us know at support@morphcraft.com and we will notify you when it becomes available.

Credits are consumed when you run an AI transformation. Each transformation type has a fixed credit cost. Credits are deducted at the start of a job and automatically refunded if the job fails.

Generate Image: 10 credits Generate Image: 10 credits Face Swap: 15 credits Character Expressions: 30 credits Portrait Animator: 20 credits Fix Distortions: 10 credits Restore Image: 15 credits Remove Background: 5 credits Upscale Image: 10 credits Style Transfer: 15 credits AI Image Edit: 25 credits Text to Video: 30 credits Text to Video (720p): 50 credits Image to Video: 70 credits Text to Video: 25 credits Video Extension: 25 credits AI Lip Sync: 50 credits First & Last Frame: 60 credits

No. Credits never expire and carry over indefinitely.

Credits are automatically refunded to your account within seconds of a failure. You will never be charged for a job that did not complete successfully.

MorphCraft supports a range of AI-powered workflows: - Background removal - Image upscaling (2x or 4x) - Style transfer (apply artistic styles to photos) - Character expressions (generate expression variations) - AI image generation from text prompts - Face swap (photo and video) - Face and body repair - Image restoration - Text-to-video and image-to-video generation - Pose transfer

Upload any image and the AI isolates the subject and removes the background. The output is a transparent PNG. For best results use images where the subject is clearly distinct from the background. Hair, fur, and semi-transparent edges are handled automatically.

Upscaling increases your image resolution by 2x or 4x using AI. It adds realistic detail rather than just stretching pixels. Useful for improving low-resolution photos or preparing images for print.

Upload a source image and a style reference image. The AI applies the visual style (colors, textures, artistic treatment) of the reference to your source image while preserving the original composition and content.

Upload a face photo and the AI generates multiple expression variations (happy, sad, surprised, etc.) from a single input. The output is a set of images — one per expression.

Choose the Generate workflow, type a description of the image you want, and the AI creates it. Be specific with your prompt — include details about subject, setting, lighting, and style for best results.

Pose transfer takes a source image and a reference pose image, then re-renders the subject in the reference pose. Upload the subject you want to repose and a second image showing the target pose.

The "Enhance faces" option is available on several workflows and applies AI face refinement as a post-processing step. It works best on images where faces are clearly visible. If the face is very small, partially occluded, or at an extreme angle, the enhancement may have limited effect.

Fix targets face and body distortions — it detects and corrects anatomy issues like warped features or misshapen hands. Restore is for missing or damaged regions — it fills in black areas, transparent patches, or corrupted sections of an image using AI.

When you upload an image to Restore, the AI automatically scans for black pixels and transparent regions. These areas are treated as damaged and filled in with AI-generated content that matches the surrounding context. You can also provide a manual mask for more control.

Large damaged regions are harder to fill convincingly. Try providing a descriptive prompt that matches the content of your image — this guides the AI toward a more consistent result. Lowering the denoise strength can also help the repaired area blend more naturally with the original.

Video generation typically takes 1 to 5 minutes depending on the model and output length. Longer or higher-resolution videos take more time. You can close the tab and come back — your job continues processing in the cloud.

All generated videos are delivered as MP4 files.

Text-to-video generates a video entirely from a text description. Image-to-video takes a still image as a starting frame and animates it based on your prompt. Image-to-video produces more predictable results because the AI has a visual reference to work from.

Be specific in your prompt about the type of motion you want (e.g. "camera slowly pans left" or "person walks forward"). Avoid very abstract or conflicting prompts. Using a higher-quality source image for image-to-video also helps produce sharper results.

Most images work well as a starting frame for image-to-video. The best results come from clear, well-composed photos with a distinct subject. Very abstract images, heavily stylized artwork, or images with text overlays may produce unpredictable results.

Use a clear, front-facing photo with good lighting. Avoid sunglasses, heavy shadows, or partial faces. JPG and PNG are supported.

Use a high-resolution source image where the face is well-lit and facing the camera. The face in the target image should also be clearly visible and roughly the same angle. Extreme angle differences between source and target faces reduce accuracy.

Face Swap works on still images. Video Face Swap applies the face swap to every frame of a video, maintaining consistency across the clip. Video face swap costs more credits because it processes many frames.

Yes — if you save a face to your account it is stored securely and only used when you explicitly request it. You can delete saved faces at any time from your account settings. Deletion is immediate and permanent.

You may only use faces you have permission to use. By uploading a face image you certify the face is yours or you have the subject's explicit consent. Creating non-consensual deepfakes violates our Terms of Service and may be illegal.

Batch processing lets you apply the same transformation to multiple images at once. Upload 2 to 50 images, configure the settings once, and all images are processed in parallel. Supported workflows: background removal, upscaling, face swap, and style transfer.

Each image in a batch costs the same as a single-image job for that workflow. Credits are deducted upfront for the full batch. If individual images within the batch fail, those credits are refunded automatically.

Yes. You own the content you create with MorphCraft. We claim no intellectual property rights over your generated images or videos.

Yes, subject to your responsibility to ensure the content does not infringe third-party rights (e.g. copyright, right of publicity). You are solely responsible for the lawful use of your generated content.

Generated images are stored for 30 days from creation, then automatically deleted. Download anything you want to keep.

We accept all major credit and debit cards via Stripe. No PayPal at this time.

Credits are non-refundable except in the case of technical failures on our end, where they are automatically returned. If you believe you were incorrectly charged, contact us at billing@morphcraft.com.

Yes — every new account receives 250 free credits on signup. No credit card required to start.

No. We do not sell personal data to any third party.

No. Your uploaded images and generated outputs are never used to train AI models.

Email us at privacy@morphcraft.com with your account email and we will delete your account and all associated data within 30 days.