关键词句AI作画

最新推荐文章于 2024-06-13 12:10:04 发布

AI Studio

最新推荐文章于 2024-06-13 12:10:04 发布

阅读量567

点赞数

文章标签：深度学习人工智能 python ffmpeg 开发语言

关键词句AI作画（基于文心大模型）

有一段优美的句子或者一个词，想配个图发朋友圈却找不到适合语境的图片？
想通过图片更好感受理解句子或词语的精妙表达？
中华文化源远流长，博大精深
把句词交给我们
我们给您一幅幅优美的“关键画”
快来跟着我们做起来吧！画儿在出口等着您呢！

AI作画示例：

在这里插入图片描述

第一步，安装wenxin_api第三方库（项目核心）

!pip install wenxin_api

Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
Requirement already satisfied: wenxin_api in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (0.0.5.1)
Requirement already satisfied: requests>=2.20 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from wenxin_api) (2.24.0)
Requirement already satisfied: tqdm in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from wenxin_api) (4.27.0)
Requirement already satisfied: idna<3,>=2.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests>=2.20->wenxin_api) (2.8)
Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests>=2.20->wenxin_api) (2019.9.11)
Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests>=2.20->wenxin_api) (1.25.6)
Requirement already satisfied: chardet<4,>=3.0.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests>=2.20->wenxin_api) (3.0.4)

[1m[[0m[34;49mnotice[0m[1;39;49m][0m[39;49m A new release of pip available: [0m[31;49m22.1.2[0m[39;49m -> [0m[32;49m22.3[0m
[1m[[0m[34;49mnotice[0m[1;39;49m][0m[39;49m To update, run: [0m[32;49mpip install --upgrade pip[0m

第二步，获取API Key

API key获取地址：https://wenxin.baidu.com/moduleApi/key

第三步，输入您的API Key 与 Secert Key

例：

API=“1314KFCVW50”
Secret=“2022KFCVW50”

API=""#请在此输入您的API Key
Secret=""#请在此输入您的Secret Key

第四步，输入需要您要作画的句子或者词语

例：data=“秋天的果园更是硕果累累，一派好景象。红艳艳的苹果像一盏盏红灯笼挂在枝头”

data="秋天的果园更是硕果累累，一派好景象。红艳艳的苹果像一盏盏红灯笼挂在枝头。"#请在双引号之间填写您的句词

第五步，获取关键词句

import wenxin_api 
from wenxin_api.tasks.summarization import Summarization
wenxin_api.ak = API
wenxin_api.sk = Secret
input_dict = {
    "text": f"文章：{data}\n标题：",
    "seq_len": 512,
    "topp": 0.3,
    "penalty_score": 1.0,
    "min_dec_len": 10,
    "is_unidirectional": 0,
    "task_prompt": "Summarization"
}
rst = Summarization.create(**input_dict)
print(rst)

2022-10-26 21:06:41,218 - model 1: starts writing
{'result': '秋天的果园更是硕果累累'}

第六步，根据关键词创作画儿

可供选择的style:古风、油画、水彩画、卡通画、二次元、浮世绘、蒸汽波艺术、low poly、像素风格、概念艺术、未来主义、赛博朋克、写实风格、洛丽塔风格、巴洛克风格、超现实主义等等（其他style选择详情请前往https://wenxin.baidu.com/wenxin/docs#Pl6llwf92）
关于resolution(尺寸），我们目前提供：10241024方图、10241536长图、1536*1024横图供您选择

import wenxin_api 
from wenxin_api.tasks.text_to_image import TextToImage
wenxin_api.ak = API
wenxin_api.sk = Secret
input_dict = {
    "text": rst,
    "style": "油画",#此处可选择您想要的画的风格（请参照上方介绍）
    "resolution": "1536*1024"#此处可更改画的尺寸，我们目前提供：1024*1024方图、1024*1536长图、1536*1024横图 供您选择
}
rst = TextToImage.create(**input_dict)
print(rst)
print("此处上方输出的每一个网址都对应一张画，可供您以网页链接的形式分享给您的好友")

2022-10-26 21:07:19,191 - model is painting now!, taskId: 7044638, waiting: 30s
2022-10-26 21:07:39,744 - model is painting now!, taskId: 7044638, waiting: 30s
{'imgUrls': ['https://wenxin.baidu.com/younger/file/ERNIE-ViLG/c004a839516cd50676c53ea1830bd848ex', 'https://wenxin.baidu.com/younger/file/ERNIE-ViLG/c004a839516cd50676c53ea1830bd848i4', 'https://wenxin.baidu.com/younger/file/ERNIE-ViLG/c004a839516cd50676c53ea1830bd8485q', 'https://wenxin.baidu.com/younger/file/ERNIE-ViLG/c004a839516cd50676c53ea1830bd84830', 'https://wenxin.baidu.com/younger/file/ERNIE-ViLG/c004a839516cd50676c53ea1830bd848v9', 'https://wenxin.baidu.com/younger/file/ERNIE-ViLG/c004a839516cd50676c53ea1830bd848a2']}
此处上方输出的每一个网址都对应一张画，可供您以网页链接的形式分享给您的好友

第七步，提取网页图片并显示

import requests
import os
from PIL import Image
import matplotlib.pyplot as plt
%matplotlib inline

# 生成存图的目录
def init_mkdir():
    data_path = 'output-img'
    if not os.path.exists(data_path): # 判断文件夹是否存在
        os.makedirs(data_path) # 不存在则新建文件夹

init_mkdir()

# 展示生成的图片并保存
for i in rst['imgUrls']:
    #使用requests直接get 下载图片使用
    r = requests.get(i)
    ii = i.split('/')[-1]+'.png'
    out_ii = 'output-img/' + ii
    with open(out_ii, 'wb') as f:
        f.write(r.content)
    f.close()
    img_ = Image.open(out_ii)
    plt.show()
    plt.imshow(img_)