人脸情感、情绪识别

准确识别包括 开心、哀伤、生气 在内的 9 种人脸表情情绪 立即体验

功能介绍

基于强大的深度学习与人脸识别算法,竹间智能实现了对人脸表情情绪及相关属性的识别分析。 该技术已经实现了对包括开心、哀伤、生气、困惑在内的 9 种人脸表情情绪识别。结合文字及语音识别,竹间的多模态情感识别已经实现了对 22 种人类情绪的精确识别。除此之外,还能实现包括年龄、性别、颜值、疲劳度、肤色、发型、妆容等 22 种人脸属性。

人脸情绪识别演示

上传一张本地图片,即刻体验竹间智能强大的人脸情绪识别能力
上传图片
[
    {
        "position": {
            "center": {
                "y": 0.78375,
                "x": 0.641121
            },
            "size": {
                "width": 0.160748,
                "height": 0.2925
            }
        },
        "face_id": 0,
        "emotions": {
            "angry": 0,
            "confused": 0,
            "sad": 0.000001,
            "neutral": 0.000102,
            "disgust": 0,
            "arousal": 0.718417,
            "surprise": 0,
            "fear": 0,
            "valence": 0.628808,
            "contempt": 0.000001,
            "happy": 0.999896
        },
        "attribute": {
            "eyeglass": 0,
            "beard_style": "no_beard",
            "gender": 1,
            "age": 26,
            "young": 1,
            "hair_color": "other",
            "hair_style": "other",
            "attractive": 0.73,
            "smile": 1
        }
    },
    {
        "position": {
            "center": {
                "y": 0.70125,
                "x": 0.819626
            },
            "size": {
                "width": 0.13271,
                "height": 0.2375
            }
        },
        "face_id": 1,
        "emotions": {
            "angry": 0,
            "confused": 0,
            "sad": 0,
            "neutral": 0,
            "disgust": 0,
            "arousal": 0.695723,
            "surprise": 0,
            "fear": 0,
            "valence": 0.685882,
            "contempt": 0,
            "happy": 1
        },
        "attribute": {
            "eyeglass": 0,
            "beard_style": "no_beard",
            "gender": 1,
            "age": 34,
            "young": 1,
            "hair_color": "black_hair",
            "hair_style": "other",
            "attractive": 0.48,
            "smile": 1
        }
    },
    {
        "position": {
            "center": {
                "y": 0.1725,
                "x": 0.241121
            },
            "size": {
                "width": 0.097196,
                "height": 0.205
            }
        },
        "face_id": 2,
        "emotions": {
            "angry": 0,
            "confused": 0,
            "sad": 0,
            "neutral": 0.000001,
            "disgust": 0.000001,
            "arousal": 0.617636,
            "surprise": 0,
            "fear": 0,
            "valence": 0.543296,
            "contempt": 0,
            "happy": 0.999999
        },
        "attribute": {
            "eyeglass": 0,
            "beard_style": "no_beard",
            "gender": 1,
            "age": 24,
            "young": 1,
            "hair_color": "other",
            "hair_style": "straight_hair",
            "attractive": 0.56,
            "smile": 1
        }
    },
    {
        "position": {
            "center": {
                "y": 0.21,
                "x": 0.48785
            },
            "size": {
                "width": 0.100935,
                "height": 0.19
            }
        },
        "face_id": 3,
        "emotions": {
            "angry": 0.000028,
            "confused": 0.001896,
            "sad": 0.000126,
            "neutral": 0.017136,
            "disgust": 0.000054,
            "arousal": 0.718403,
            "surprise": 0.000453,
            "fear": 0.000239,
            "valence": 0.817017,
            "contempt": 0.000649,
            "happy": 0.979418
        },
        "attribute": {
            "eyeglass": 1,
            "beard_style": "no_beard",
            "gender": 0,
            "age": 38,
            "young": 0,
            "hair_color": "other",
            "hair_style": "other",
            "attractive": 0.46,
            "smile": 0
        }
    },
    {
        "position": {
            "center": {
                "y": 0.35125,
                "x": 0.691589
            },
            "size": {
                "width": 0.097196,
                "height": 0.1875
            }
        },
        "face_id": 4,
        "emotions": {
            "angry": 0,
            "confused": 0,
            "sad": 0,
            "neutral": 0,
            "disgust": 0,
            "arousal": 0.800219,
            "surprise": 0,
            "fear": 0,
            "valence": 0.81731,
            "contempt": 0,
            "happy": 0.999999
        },
        "attribute": {
            "eyeglass": 1,
            "beard_style": "no_beard",
            "gender": 0,
            "age": 33,
            "young": 0,
            "hair_color": "black_hair",
            "hair_style": "straight_hair",
            "attractive": 0.34,
            "smile": 1
        }
    },
    {
        "position": {
            "center": {
                "y": 0.32,
                "x": 0.856075
            },
            "size": {
                "width": 0.08972,
                "height": 0.13
            }
        },
        "face_id": 5,
        "emotions": {
            "angry": 0.000155,
            "confused": 0.00529,
            "sad": 0.001343,
            "neutral": 0.616573,
            "disgust": 0.000051,
            "arousal": 0.618072,
            "surprise": 0.000062,
            "fear": 0.000099,
            "valence": 0.506505,
            "contempt": 0.00003,
            "happy": 0.376397
        },
        "attribute": {
            "eyeglass": 0,
            "beard_style": "no_beard",
            "gender": 0,
            "age": 28,
            "young": 1,
            "hair_color": "black_hair",
            "hair_style": "straight_hair",
            "attractive": 0.57,
            "smile": 0
        }
    }
]
                                    
  • 情绪
    开心
  • 性别
  • 年龄
    26
  • 颜值
    73
  • 发色
    其他
欢迎体验竹间开放平台API  点击立刻体验

技术特色

  • 独家的情绪识别技术

    能够准确识别9种人脸表情情绪。结合文字及语音等多模态情感情绪识别,可实现对22种人类情绪的精确识别。
  • 丰富的识别维度

    除情绪识别外,竹间人脸识别技术还支持对年龄、性别、颜值、疲劳度、肤色、发型、妆容等 22 种丰富的人脸属性识别。
  • 广泛的应用场景

    除帮助企业深入了解用户外,此项技术还能应用于对话机器人、服务机器人、智能汽车驾驶系统、IoT、广告效果监测等广泛的应用场景。
  • 更方便的部署方式

    竹间人脸情绪识别端口支持灵活的部署方式,允许云端或本地的多种设备接入,而且能快速得到识别结果和反馈信息。

应用场景