Avatar SDK Standalone Application  2.0.8
Head 2.0 Avatars
Parameters Description

Computation parameters are provided in JSON format that contains the following section: pipeline, pipeline_subtype, blendshapes, haircuts, model_info, avatar_modifications, additional_textures.
The full list of parameters can be found in the parameters_samples/bust_mobile_subtype_parameters.json and parameters_samples/head_mobile_subtype_parameters.json sample files.

Pipeline and Pipeline Subtype

pipeline should always be set to head_2.0 as the current version of the Standalone application generates only Head 2.0 avatars.
Possible values for the pipeline_subtype are bust/mobile and head/mobile.

Blendshapes

Contains blendshapes sets names that should be generated for the avatar.
Blendshapes are stored in the *.bin* file format. It contains consequent three coordinates (X,Y,Z) of vertex deltas as float values. In case the avatar is generated in the FBX format, blendshapes are embedded into this FBX.

Haircuts

Contains haircut names that should be generated for the avatar. Each haircut is represented as a separate mesh.

Model Info

The model_info section contains flags specifying which meta-information should be computed for the avatar. The generated meta-information is stored in the model.json file in the output directory. More information about model_info parameters can be found in our Web Api Documentation.

"model_info":
[
"gender",
"race",
"age",
"hair_color",
"skin_color",
"eye_sclera_color",
"eye_iris_color",
"predict_haircut",
"facial_landmarks_68"
]

gender: predict a person's gender from a submitted photo. Possible values are male and female.
race: predict a person race from the submitted photo. Currently, three races are predicted (asian, black and white) with certain confidence for each race in the range [0, 1] inclusive.
age: classify a person's age from a submitted photo. Currently possible values are child and not_child.
hair_color: compute the average haircut color from a submitted photo.
skin_color: compute the average skin color from a submitted photo.
eye_sclera_color: compute the average eye sclera color from a submitted photo.
eye_iris_color: compute the average eye iris color from a submitted photo.
predict_haircut: predict which artificial haircut matches best to the submitted photo.
facial_landmarks_68: compute facial landmarks. Landmark coordinates are represented as a flat JSON array with X, Y, Z float components sequence for each landmark, or NaN (for each component) if no landmark is detected. Available only for the "head/mobile" subtype.

Avatar Modifications

Parameters configure the avatar appearance. More information about avatar_modifications parameters can be found in our Web Api Documentation

"avatar_modifications":
{
"curved_bottom" : true,
"remove_smile": true,
"remove_glasses": true,
"enhance_lighting" : true,
"remove_stubble" : false,
"parametric_eyes_texture_v2" : false,
"parametric_eyes_texture" : false,
"add_glare" : false,
"add_eyelid_shadow" : false,
"eye_iris_color" : { "blue" : 82, "green" : 76, "red" : 73 },
"eye_sclera_color" : { "blue" : 191, "green" : 192, "red" : 198 },
"hair_color" : { "blue" : 67, "green" : 94, "red" : 128 },
"teeth_color" : { "blue" : 200, "green" : 200, "red" : 200 },
"texture_size" : { "height" : 2048, "width" : 2048 },
"generated_haircut_texture_size" : { "height" : 1024, "width" : 1024 },
"generated_haircut_faces_count" : 5000
}

curved_bottom: make the bottom of the model bust slightly curved. Available only for the "bust/mobile" subtype.
remove_smile: attempt to remove a smile on the submitted photo before the actual avatar computation starts.
remove_glasses: attempt to remove glasses on the submitted photo before the actual avatar computation starts.
enhance_lighting: attempt to make lighting more uniform.
remove_stubble: attempt to remove a stubble on the submitted photo before the actual avatar computation start.
parametric_eyes_texture_v2: replace the eye texture from the submitted photo with generated one with sclera and iris colors to match the photo version.
parametric_eyes_texture: replace the eye texture from the submitted photo with the generated one with sclera and iris colors to match the photo version.
add_glare: add a glare directly into the parametric eye texture. Works only when parametric_eyes_texture is set. Enabled by default.
add_eyelid_shadow: add an eyelid shadow on top of the eye texture. Works only when parametric_eyes_texture is set.
eye_iris_color: recolor the eye iris directly on a model texture. Works only when parametric_eyes_texture is set.
eye_sclera_color: recolor eye sclera directly on a model texture. Works only when parametric_eyes_texture is set.
hair_color: recolor generated haircut directly on a haircut texture.
teeth_color: recolor teeth directly on a model texture.
texture_size: resize the model texture to the specified size. The original texture size is 2048x2048 px, thus setting a size greater than this value will not result in improved texture quality.
generated_haircut_texture_size: resize generated haircut texture to the specified size. The original haircut texture size is 1024x1024 px, thus setting a size greater than this value will not result in improved texture quality.
generated_haircut_faces_count: resample generated haircut mesh to have specified faces count.

Additional Textures

A list with additional textures will be generated for the avatar. More information about additional_textures parameters can be found in our Web Api Documentation.

lips_mask: mask in PNG format that shows the location of lips.
metallic_map: metalness map in PNG format.
roughness_map: roughness map in PNG format.
normal_map: normal map in PNG format.