Giter Site home page Giter Site logo

suchka-ai's Introduction

Suchka-AI

目标数据结构 Movie

  • faces
  • poses
  • objects
  • image_features

Data Source

机器学习的数据源通过自动爬去这些数据源,数据源的数据结构如下 所有图片都存放在同一个文件夹下, 所有图片都转换成 jpg 格式,分辨率以最小边长512pix缩放。 所有的描述信息存放在图片的metadata信息中 例如: md5(url).jpg

自定义标签:

  • imageDescription: captions
  • SKA_DESCRIPTION: image captions (text label)
  • SKA_TAGS: csv tags
  • SKA_CATEGORIES: csv category names

Picture database

Crawler

scrapy crawler ./crawler/pornpics.py

AI 学习

判断演员

判断视频所有关键帧,从中提取出人脸做特征提取。 然后把所有的人脸放在一个集合内做聚类,由这个逻辑判断主演的面部特征码 DeepFace 是一个比较好的统一类库

问题

男演员问题

通常男演员的面部特别特征特别稀疏,一个图集中只有很少的出境 等于要把所有图片集中的脸识别出来,按图集分类,比如图集中出现了3个独立的面孔 然后找出各个图集中的共同脸形成这个模特的最原始人脸信息

当已经形成了模特的人脸信息后,其他照片和这个信息做比对

Visualizaion

pyvis

多标签分类任务

特征提取

特征提取用ResNet101

分类头

参考文章 https://zhuanlan.zhihu.com/p/107737824

tensorflow GPU

source $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh

Useful checker

// check if duplicate records, expect to be empty output
db.gallery.aggregate([ { $group: { _id: "$gid", count: { $sum: 1 } } }, { $match: { count: { $gt: 1 } } }]);
db.channel.aggregate([ { $group: { _id: "$name", count: { $sum: 1 } } }, { $match: { count: { $gt: 1 } } }]);
db.model.aggregate([ { $group: { _id: "$name", count: { $sum: 1 } } }, { $match: { count: { $gt: 1 } } }]);
db.tag.aggregate([ { $group: { _id: "$name", count: { $sum: 1 } } }, { $match: { count: { $gt: 1 } } }]);
db.face.aggregate([ { $group: { _id: {$concat:["$name","$source"]}, count: { $sum: 1 } } }, { $match: { count: { $gt: 1 } } }]);

// check data records for each collection
db.getCollectionNames().forEach((collectionName) => {
  const count = db.getCollection(collectionName).countDocuments();
  print(`Collection: ${collectionName}, Count: ${count}`);
});
// find the most popular tags
db.tag.aggregate([{$project: {name: "$name" ,g_count: {$size: "$galleries"}}}, {$sort: {g_count:-1}}])

backup database

docker exec -it mongo mongodump -d suchka-ai  -o /data/backup
docker cp mongo:/data/backup/suchka-ai /backup/suchka-ai/db
docker exec -it mongo rm -rf /data/backup/suchka-ai
(cd /backup/suchka-ai/db; tar czvf suchka-ai.tar.gz  ./suchka-ai; rm -rf ./suchka-ai)

suchka-ai's People

Contributors

hellojixian avatar

Stargazers

 avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.