# Avatar 101: Negative examples URL: https://www.atmee.ai/docs/v2/guides/negative-examples > The portraits that animate badly, and what to use instead. Style and framing are never the problem: Atmee V2 takes half-body, full-body and even squatting shots. An avatar is only as good as the portrait it's built from. This page lists only what goes wrong, so you can rule it out. Anything not on it is fine. Every example below is the same invented head-and-shoulders photo with one problem added, so that the problem is the only thing that differs. That framing is not what to aim for. Close-ups, half-body and full-body shots, standing, sitting or even squatting, all work on Atmee V2, and so do photos, paintings, illustrations and cartoons. ## Framing limits on each model The two models differ most in the framing they accept. | | Atmee V2 | Atmee V1 | | ----------- | --------------------------------------------------------------------------------------------------------- | ------------------------------------- | | Framing | Any: close-up, head and shoulders, half-body or full-body, standing, sitting or squatting | Head and shoulders, facing the camera | | Face size | Wider than about a fifth of the avatar's frame | At least about a third of the frame | | Frame shape | Square (512 × 512) by default; portrait (384 × 672) or landscape (672 × 384) where your account offers it | Square | | Expression | Relaxed, with a visible mouth and jaw | Relaxed, lips slightly parted | On Atmee V2 a wider shot shows more of the body moving as the avatar talks, so there is no need to crop in close. Choose the frame shape when you create the avatar (`resolution` in the [API](/docs/v2/api/create-avatar)); changing it later builds a new version. Both models take jpg, png or webp of at most 10 MB, at least 512 pixels on the short side. See [Models](/docs/v2/getting-started/models) for the rest of the differences. ## How each problem is marked * **Won't work:** the avatar can't be built well from it. * **Likely to fail:** it usually animates badly. * **May work:** try it, and compare the result. ## Framing Atmee scales the photo until it fills the avatar's frame, then keeps the middle. A wide photo for a square avatar loses its sides, and a person far from the camera ends up a few dozen pixels wide. Use a photo shaped like the frame, or one with the face in the middle and big enough. - **Part of the head is cut off by the photo's edge** (Won't work). Use a photo with the whole head and some space above it. - **The face sits near the edge of a wide photo** (Won't work). Use a photo shaped like the avatar (square by default), or one with the face in the middle. - **The person is small in a large frame** (Likely to fail). Use a closer photo, or one where your face takes up more of it. - **Only the face, with no hair outline or neck** (Likely to fail). Use a photo that shows at least your whole head and neck; half-body and full-body shots work too on Atmee V2. - **The head or camera is tilted** (May work). Use a photo taken straight on, with your head level. ## Mouth and face Lip sync needs a clear view of the mouth and jaw. - **Something covers the mouth** (Won't work). Use a photo where nothing is in front of your mouth. - **An object in front of the chin or lips** (Likely to fail). Use a photo with nothing near your mouth. - **Hair covers part of the face** (Likely to fail). Use a photo with your hair away from your face. - **A beard or moustache hides the lips** (Likely to fail). If you can, use a photo where your lips are visible; otherwise try it, and compare the takes carefully. - **Sunglasses or glasses with strong glare** (May work). Use a photo without sunglasses, or with clear glasses and no glare. - **A hat shades the eyes** (May work). Use a photo without the hat, or with your eyes in the light. ## Pose and expression The face in the photo is the avatar's resting face for every conversation. - **A strong expression** (Likely to fail). Use a photo with a relaxed face and a slight smile. - **Tongue out or a pulled face** (Likely to fail). Use a photo with a relaxed face and a slight smile. - **Eyes looking away from the camera** (May work). Use a photo where you look straight into the camera. ## Quality - **The photo is too small or upscaled** (Likely to fail). Use the original photo file, at least the avatar's frame size (512×512 by default). - **The photo is out of focus or has motion blur** (Likely to fail). Use a sharp photo; hold the camera still. - **Heavy JPEG compression** (May work). Use the original photo, not one saved from a chat app or social media. - **A photo of a screen or a printed photo** (Likely to fail). Use the original digital photo. ## Light - **The face is washed out** (Likely to fail). Use a photo in even, softer light. - **Backlit: a bright background and a dark face** (Likely to fail). Use a photo with the light in front of you, not behind. - **Strong coloured light** (May work). Use a photo in plain daylight or white light. - **A hard shadow across the face** (Likely to fail). Use a photo with even light on both sides of your face. ## Other content - **A collage, grid or before/after image** (Won't work). Use a single photo. - **Filters or AR stickers on the face** (Likely to fail). Use the photo without filters or stickers. - **Another face in the photo** (Likely to fail). Use a photo with only you in it. A few more photos animate badly too, with no example above: a face in profile, a mask, helmet or mascot head that hides the mouth, a close wide-angle selfie that distorts the face, text or a logo over the face, and a background the same colour as the hair. ## Voice samples that fail A voice sample is 10 to 60 seconds of one person speaking clearly. Atmee checks it for clean, audible speech, so these fail or sound wrong: * shorter than 10 seconds or longer than 60, * silence, a tone, or too little speech, * music under the voice, * more than one voice. Formats and limits are in [Create an avatar](/docs/v2/api/create-avatar). ## If the result isn't right * **Atmee V2:** capture more takes of the avatar and keep the one you like best. Each take is a new short clip from the same portrait and voice. * **Either model:** replace the portrait or the voice. The avatar builds again from the new one. - [Create an avatar](/docs/v2/api/create-avatar): Build one through the API, with a portrait, a voice and a persona. - [Models](/docs/v2/getting-started/models): Atmee V2 and Atmee V1 side by side.