Unity & FACEGOOD Audio2Face 通过音频驱动面部BlendShape

简介: Unity & FACEGOOD Audio2Face 通过音频驱动面部BlendShape

 目录

Audio2Face简介

在Unity中应用


Audio2Face简介

在元宇宙的热潮下,为了让AI数字人渗透到更多的领域中,FACEGOOD已经将语音驱动口型的算法技术开源,开源地址:

https://github.com/FACEGOOD/FACEGOOD-Audio2Face

image.gif

该技术可以实时将音频数据转换为驱动数字人面部BlendShape的权重数据,不同于ARKit中的52个BlendShape,它的数量多达116个,我们可以通过对应关系得到相应的数值,对应关系如下:

ARKit Voice2Face
eyeBlinkLeft eye_blink2_l
eyeLookDownLeft eye_lookDown2_l
eyeLookInLeft eye_lookRight_l
eyeLookOutLeft eye_lookLeft_l
eyeLookUpLeft eye_lookUp_l
eyeSquintLeft eye_shutTight_l
eyeWideLeft max(eye_downLidRaise_l,eye_upLidRaise_l)
eyeBlinkRight eye_blink2_r
eyeLookDownRight eye_lookDown2_r
eyeLookInRight eye_lookRight_r
eyeLookOutRight eye_lookLeft_r
eyeLookUpRight eye_lookUp_r
eyeSquintRight eye_shutTight_r
eyeWideRight max(eye_downLidRaise_r,eye_upLidRaise_r)
jawForward jaw_thrust_c
jawLeft jaw_sideways_l
jawRight jaw_sideways_r
jawOpen mouth_stretch_c
mouthClose mouth_chew_c
mouthFunnel max(mouth_funnel_dl,mouth_funnel_dr,mouth_funnel_ul,mouth_funnel_ur)
mouthPucker max(mouth_pucker_l,mouth_pucker_r)
mouthLeft mouth_sideways_l
mouthRight mouth_sideways_r
mouthSmileLeft mouth_lipCornerPull_l
mouthSmileRight mouth_lipCornerPull_r
mouthFrownLeft max(mouth_lipCornerDepress_l,mouth_lipCornerDepressFix_l)
mouthFrownRight max(mouth_lipCornerDepress_r,mouth_lipCornerDepressFix_r)
mouthDimpleLeft mouth_dimple_l
mouthDimpleRight mouth_dimple_r
mouthStretchLeft mouth_lipStretch_l
mouthStretchRight mouth_lipStretch_r
mouthRollLower max(mouth_suck_dl,mouth_suck_dr)
mouthRollUpper max(mouth_suck_ul,mouth_suck_ur)
mouthShrugLower mouth_chinRaise_d
mouthShrugUpper mouth_chinRaise_u
mouthPressLeft mouth_press_l
mouthPressRight mouth_press_r
mouthLowerDownLeft mouth_lowerLipDepress_l
mouthLowerDownRight mouth_lowerLipDepress_r
mouthUpperUpLeft mouth_upperLipRaise_l
mouthUpperUpRight mouth_upperLipRaise_r
browDownleft brow_lower_l
browDownRight brow_lower_r
browInnerUp brow_raise_c
browOuterUpLeft brow_raise_l
browOuterUpRight brow_raise_r
cheekPuff max(cheek_puff_l,cheek_puff_r)
cheekSquintLeft cheek_up
cheekSquintRight cheek_up
noseSneerLeft nose_out_l
noseSneerRight nose_out_r
tongueOut

生产的数据结果如下图所示,可见是116个取值范围为-1~1的小数:

image.gif

这116个数值依次对应下面116个BlendShape名称:

brow_lower_l
tongue_Scale__X
tongue_Scale_Y
tongue_Scale__Y
tongue_Scale_Z
tongue_Scale__Z
nose_out_l
nose_out_r
tongue_u
tongue_u_u
brow_raise_d
cheek_suck_r
mouth_stretch_u
tongue_u_d
tooth_d_d
tongue_d
tooth_r
tooth_d_u
cheek_UP
eye_blink1_l
eye_blink1_r
eye_blink2_l
eye_blink2_r
eye_lidTight_l
eye_lidTight_r
eye_shutTight_l
eye_shutTight_r
brow_lower_r
eye_upperLidRaise_l
eye_upperLidRaise_r
eye_downLidRaise_l
eye_downLidRaise_r
jaw_sideways_l
jaw_sideways_r
jaw_thrust_c
mouth_chew_c
mouth_chinRaise_d
mouth_chinRaise_u
brow_raise_c
mouth_dimple_l
mouth_dimple_r
mouth_funnel_dl
mouth_funnel_dr
mouth_funnel_ul
mouth_funnel_ur
mouth_lipCornerDepressFix_l
mouth_lipCornerDepressFix_r
mouth_lipCornerDepress_l
mouth_lipCornerDepress_r
brow_raise_l
mouth_lipCornerPullOpen_l
mouth_lipCornerPullOpen_r
mouth_lipCornerPull_l
mouth_lipCornerPull_r
mouth_lipStretchOpen_l
mouth_lipStretchOpen_r
mouth_lipStretch_l
mouth_lipStretch_r
mouth_lowerLipDepress_l
mouth_lowerLipDepress_r
brow_raise_r
mouth_lowerLipProtrude_c
mouth_oh_c
mouth_oo_c
mouth_pressFix_c
mouth_press_l
mouth_press_r
mouth_pucker_l
mouth_pucker_r
mouth_screamFix_c
mouth_sideways_l
cheek_puff_l
mouth_sideways_r
mouth_stretch_c
mouth_suck_dl
mouth_suck_dr
mouth_suck_ul
mouth_suck_ur
mouth_upperLipRaise_l
mouth_upperLipRaise_r
nose_wrinkle_l
nose_wrinkle_r
cheek_puff_r
tooth_l
eye_lookDown1_l
eye_lookDown2_l
eye_lookLeft_l
eye_lookRight_l
eye_lookUp_l
eye_lookDown1_r
eye_lookDown2_r
eye_lookLeft_r
eye_lookRight_r
cheek_raise_l
eye_lookUp_r
tongue_Rot_1X
tongue_Rot__1X
tongue_Rot_2X
tongue_Rot__2X
tongue_Rot_3X
tongue_Rot__3X
tongue_Rot_1Y
tongue_Rot__1Y
tongue_Rot_2Y
cheek_raise_r
tongue_Rot__2Y
tongue_Rot_3Y
tongue_Rot__3Y
tongue_Rot_1Z
tongue_Rot__1Z
tongue_Rot_2Z
tongue_Rot__2Z
tongue_Rot_3Z
tongue_Rot__3Z
tongue_Scale_X
cheek_suck_l

image.gif

在Unity中应用

可以用过构建python服务,Unity客户端开启麦克风录制音频,将音频数据发送给python服务端,服务端转换为驱动BlendShape的权重数据后,返回给Unity客户端进行驱动。需要注意的是Unity中BlendShape的权重范围并不是[-1,1],因此需要进行映射。

image.gif

例如:

//将[-1,1]映射到[-100,100]
private float Remap(float v)
{
    return v * 100f;
}

image.gif

下面是一段测试音频产生的bs权重数据文件,每一行包含116个权重数值,我们拿来进行测试,将其放到StreamingAssets文件夹下。

image.gif

测试模型:

image.gif

测试代码:

using System.IO;
using System.Collections;
using System.Collections.Generic;
using UnityEngine;
public class TEST : MonoBehaviour
{
    private Coroutine coroutine;
    private SkinnedMeshRenderer smr;
    private readonly List<List<float>> valueList = new List<List<float>>();
    private IEnumerator Start()
    {
        smr = GetComponent<SkinnedMeshRenderer>();
        string path = Path.Combine(Application.streamingAssetsPath, "weight.txt");
        using (StreamReader streamReader = new StreamReader(path))
        {
            string content;
            while ((content = streamReader.ReadLine()) != null)
            {
                List<float> list = new List<float>();
                content = content.Trim();
                string[] splitArray = content.Split(' ');
                for (int i = 0; i < splitArray.Length; i++)
                {
                    float.TryParse(splitArray[i], out float result);
                    list.Add(result);
                }
                valueList.Add(list);
                yield return null;
            }
        }
    }
    private IEnumerator ExecuteCoroutine()
    {
        for (int i = 0; i < valueList.Count; i++)
        {
            List<float> list = valueList[i];
            smr.SetBlendShapeWeight(0, Remap(list[49]));    //brow_raise_l
            smr.SetBlendShapeWeight(1, Remap(list[60]));    //brow_raise_r
            smr.SetBlendShapeWeight(2, Remap(list[25]));    //eye_shutTight_l
            smr.SetBlendShapeWeight(3, Remap(list[26]));    //eye_shutTight_r
            smr.SetBlendShapeWeight(4, Remap(list[87]));    //eye_lookRight_l
            smr.SetBlendShapeWeight(5, Remap(list[86]));    //eye_lookLeft_l    
            smr.SetBlendShapeWeight(6, Remap(list[92]));    //eye_lookRight_r
            smr.SetBlendShapeWeight(7, Remap(list[91]));    //eye_lookLeft_r
            smr.SetBlendShapeWeight(8, Remap(list[88]));    //eye_lookUp_l
            smr.SetBlendShapeWeight(9, Remap(list[94]));    //eye_lookUp_r
            smr.SetBlendShapeWeight(10, Remap(list[85]));   //eye_lookDown2_l
            smr.SetBlendShapeWeight(11, Remap(list[90]));   //eye_lookDown2_r
            smr.SetBlendShapeWeight(12, Mathf.Max(Remap(list[71]), Remap(list[82])));   //cheek_pull_l cheek_pull_r
            smr.SetBlendShapeWeight(13, Remap(list[18]));   //cheek_UP
            smr.SetBlendShapeWeight(14, Remap(list[18]));   //cheek_UP
            smr.SetBlendShapeWeight(15, Remap(list[6]));    //nose_out_l   
            smr.SetBlendShapeWeight(16, Remap(list[7]));    //nose_out_r
            smr.SetBlendShapeWeight(17, Remap(list[70]));   //mouth_sideways_l
            smr.SetBlendShapeWeight(18, Remap(list[72]));   //mouth_sideways_r
            smr.SetBlendShapeWeight(19, Mathf.Max(Remap(list[67]), Remap(list[68])));   //mouth_pucker_l mouth_pucker_2
            smr.SetBlendShapeWeight(20, Mathf.Max(Remap(list[41]), Remap(list[42]), Remap(list[43]), Remap(list[44])));   //mouth_funnel_dl dr ul ur
            smr.SetBlendShapeWeight(21, Remap(list[52]));   //mouth_lipCornerPull_l
            smr.SetBlendShapeWeight(22, Remap(list[53]));   //mouth_lipCornerPull_r
            smr.SetBlendShapeWeight(23, Mathf.Max(Remap(list[47]), Remap(list[45])));   //mouth_lipCornerDepress_l mouth_lipCornerDepressFix_l
            smr.SetBlendShapeWeight(24, Mathf.Max(Remap(list[48]), Remap(list[46])));   //mouth_lipCornerDepress_r mouth_lipCornerDepressFix_r
            smr.SetBlendShapeWeight(25, Remap(list[39]));   //mouth_dimple_l
            smr.SetBlendShapeWeight(26, Remap(list[40]));   //mouth_dimple_r
            smr.SetBlendShapeWeight(27, Remap(list[65]));   //mouth_press_l
            smr.SetBlendShapeWeight(28, Remap(list[66]));   //mouth_press_r
            smr.SetBlendShapeWeight(29, Remap(list[36]));   //mouth_chinRaise_d
            smr.SetBlendShapeWeight(30, Remap(list[37]));   //mouth_chinRaise_u
            smr.SetBlendShapeWeight(31, Remap(list[56]));   //mouth_lipStretch_l
            smr.SetBlendShapeWeight(32, Remap(list[57]));   //mouth_lipStretch_r
            smr.SetBlendShapeWeight(33, Remap(list[78]));   //mouth_upperLipRaise_l
            smr.SetBlendShapeWeight(34, Remap(list[79]));   //mouth_upperLipRaise_r
            smr.SetBlendShapeWeight(35, Remap(list[58]));   //mouth_lowerLipDepress_l
            smr.SetBlendShapeWeight(36, Remap(list[59]));   //mouth_lowerLipDepress_r
            smr.SetBlendShapeWeight(37, Mathf.Max(Remap(list[76]), Remap(list[77])));    //mouth_suck_ul mouth_suck_ur
            smr.SetBlendShapeWeight(38, Mathf.Max(Remap(list[74]), Remap(list[75])));    //mouth_suck_dl mouth_suck_dr
            smr.SetBlendShapeWeight(39, Remap(list[35]));   //mouth_chew_c
            smr.SetBlendShapeWeight(40, Remap(list[34]));   //jaw_thrust_c
            smr.SetBlendShapeWeight(41, Remap(list[73]));   //mouth_stretch_c
            smr.SetBlendShapeWeight(42, Remap(list[32]));   //jaw_sideways_l
            smr.SetBlendShapeWeight(43, Remap(list[33]));   //jaw_sideways_r
            smr.SetBlendShapeWeight(44, Remap(list[38]));   //brow_raise_c
            smr.SetBlendShapeWeight(45, Remap(list[22]));   //eye_blink2_r
            smr.SetBlendShapeWeight(46, Remap(list[21]));   //eye_blink2_l
            smr.SetBlendShapeWeight(47, Remap(list[0]));   //brow_lower_l
            smr.SetBlendShapeWeight(48, Remap(list[27]));   //brow_lower_r
            smr.SetBlendShapeWeight(49, Mathf.Max(Remap(list[31]), Remap(list[29])));   //eye_downLidRaise_r eye_upLidRaise_r
            smr.SetBlendShapeWeight(50, Mathf.Max(Remap(list[30]), Remap(list[28])));   //eye_downLidRaise_l eye_upLidRaise_l
            yield return new WaitForSeconds(.07f);
        }
        coroutine = null;
    }
    private float Remap(float v)
    {
        return v * 100f;
    }
    private void OnGUI()
    {
        GUI.enabled = coroutine == null;
        if (GUILayout.Button("Begin", GUILayout.Width(200f), GUILayout.Height(50f)))
        {
            coroutine = StartCoroutine(ExecuteCoroutine());
        }
        GUI.enabled = coroutine != null;
        if (GUILayout.Button("Stop", GUILayout.Width(200f), GUILayout.Height(50f)))
        {
            StopCoroutine(coroutine);
            coroutine = null;
        }
    }
}

image.gif

image.gif

目录
相关文章
|
4月前
72B、1.8B、Audio模型
72B和1.8B是两个不同的模型,具体区别如下: - 72B是一个相对较大的模型,拥有72个亿个参数,而1.8B只有180亿个参数。
104 4
|
4月前
|
机器学习/深度学习 人工智能 数据库
CV项目手势识别(二)
这里用的语言是python,后面也是一直用python,因为这种语言在人工智能领域正在被广泛地使用,同时因为其容易上手并且数据库强大而受到欢迎。
|
7月前
|
C++ 计算机视觉 Python
vs qt opencv c++图片相片查看编辑工具Image Viewer Image Editer
vs qt opencv c++图片相片查看编辑工具Image Viewer Image Editer
118 0
|
人工智能 API iOS开发
iOS MachineLearning 系列(2)—— 静态图像分析之矩形识别
本系列文章将完整的介绍iOS中Machine Learning相关技术的应用。本篇文章开始,我们将先介绍一些与Machine Learning相关的API的应用。使用这些API可以快速方便的实现很多如图像识别,分析等复杂功能,且不会增加应用安装包的体积。
222 0
|
图形学 iOS开发
Unity【Face Cap】- 关于人脸捕捉的解决方案(二)
Unity【Face Cap】- 关于人脸捕捉的解决方案(二)
670 0
Unity【Face Cap】- 关于人脸捕捉的解决方案(二)
|
图形学
Unity Shader Graph 制作Hologram全息效果
Unity Shader Graph 制作Hologram全息效果
197 0
Unity Shader Graph 制作Hologram全息效果
|
人工智能 计算机视觉
CV之FD:基于dlib、cv2库利用warpPerspective函数和_68_face_landmarks文件实现AI换脸渐变融合视频效果
CV之FD:基于dlib、cv2库利用warpPerspective函数和_68_face_landmarks文件实现AI换脸渐变融合视频效果
CV之FD:基于dlib、cv2库利用warpPerspective函数和_68_face_landmarks文件实现AI换脸渐变融合视频效果
|
并行计算 Ubuntu 数据安全/隐私保护
图片转换为 mp4 video,一项神奇的技术【CVPR 2020】
图像修复-代码环境搭建-知识总结
728 0
图片转换为 mp4  video,一项神奇的技术【CVPR 2020】
|
机器学习/深度学习 计算机视觉 Python
基于face_recognition实现人脸识别
## 简介 我们这次使用基于开源项目face_recognition库来实现人脸识别,首先介绍一下这个项目吧。 使用世界上最简单的人脸识别库从 Python 或命令行识别和操作人脸。 使用dlib使用深度学习构建的最先进的人脸识别技术构建。该模型在 Wild基准的 Labeled Faces 上的准确率为 99.38% 。 这使得我们可以直接调用这个库来进行人脸识别而不用自己编写程序进行深度学习。这是该项目的[github地址](https://github.com/ageitgey/face_recognition)
258 1