Overview & Supported Outputs
Reactive Reality can transform regular model photos into avatars (
AVATAR_2D) for Fashion Image Generation. Finished avatars will be shown in the PICTOFiT Content Management System for virtual styling.
Content Service Type:
For more information on Output Asset Type Specifications, click here.
Image Quality Specifications
JPG (JPEG) or PNG
Supported file types for
Ideal size is 500 KB or less
Images with large file size tend to affect site performance.
1500 to 2500 pixels wide
Images smaller than the recommended baseline will appear blurry or pixelated
Up to 60 MP (megapixel)
Image resolution can be calculated by multiplying the dimensions of the image and dividing the product by 1 million
Supported color mode
The images will not look good on devices if the color profile is not sRGB
Input Avatar Image Requirements
We accept front and back views of model images for
AVATAR_2Dcreation and virtual styling. Please note that front view avatars are intended to be fitted with front view
GHOST MANNEQUIN IMAGEof the garment and vice versa (back view garment images and avatars) for a realistic result.
Avatar Quality Specifications
For images showing the model’s back, the hair must be tied to a bun or placed on the side to the front.
Location & Background
Initial Garment Specifications
This section defines and illustrates the recommended garments that models can wear initially for
AVATAR_2Dcreation. This varies depending on the customer, whether they require a full (top to bottom) or partial (top or bottom) outfit.
When choosing an initial garment, it is best to refer to the
GARMENT_2DInput Specifications to check if the initial garment matches the type of clothing that will be fitted to the model (e.g. swimwear, active wear, sleeveless dresses/tops, and so on.
List of Acceptable Bra Types
List of Acceptable Panties
We recommend to photo shoot the model for full outfit presentation and virtually fit the pants.
We recommend to photo shoot the model for full outfit presentation and virtually fit the top.
Generally, the pose of a model should meet the following specifications:
The pose must be static. The model is standing in front of the camera.
The torso (from the bust to the waist) should always face the camera, resembling the frontal product image of the garment.
All body parts must be fully visible, showing a clear outline/shape of the human body.
Specifically, we categorize these poses depending on their accuracy when garments are fitted to them using our technology:
Straight A Pose
The head is straight and not tilted.
Extremities are symmetrical from left to right.
The arms are straight and is separated from the torso, maintaining a safe distance.
Legs are not spread out too much, maintaining a safe distance between them.
Both feet are positioned forward.
All body parts are visible and show a clear outline.
Minor Bends & Tilt
The head is slightly tilted but the neck is still visible
Arms are slightly bent. Although the left & right arms can be a bit asymmetrical, it still needs to be aligned.
Slight bend on the waist and hips making one side a bit more prominent.
All body parts are still visible and shows a clear outline.
Arms Closer to the Body
The upper arms are not completely covered by the chest and lower arms maintain a distance between the thighs.
All body parts are still visible and show a clear outline.
Not Processable Poses
The following poses cannot be transformed into
Arms covering the torso
Arms raised upwards