QA Official

Probability Distribution Face Quality Evaluation Based on Frequency Components 2019-05-01
Original Work: 《Patch-based Probabilistic Image Quality Assessment for Face Selection and Improved Video-based Face 》 The main idea of this paper is to divide the human face into many 8×8 small blocks. The author thinks that each block represents a different part of the face, extracts the first three AC components of the block respectively, and then

Raspberry Pi TV Box Based on Xbian 2019-05-01
Take some time today to write a detailed tutorial for you.This tutorial mainly teaches you to download xbian, install, configure in Chinese, install video plug-ins and watch movies...Declaration:All the peripherals and accessories mentioned in this tutorial are available at the official online shop of the forum ( )This tutorial is based on the Windows 7 (64-bit) system. Other Windows systems are basically the same.8G SD card is recommended.The monitor is VGA interface and uses HDMI to VGA (other HDMI interface display devices are the same)keyboard is a must, mouse can not *****************************************************************************official start:1.

Some Summaries of pureMVC 2019-05-01
这次项目中的框架用到了PureMVC,坦白的讲虽然用了,但并没有很明显的体会到 它给项目带来了些什么具体的、显而易见的优势,网上有说他能够摒弃

Use Open Source Face Feature Extractor to Score Face Color Value 2019-05-01
This article is only for model application and actual combat, not color value research. The results are for entertainment and reference only. method is only for reference. Generally speaking, the larger the amount of data, the closer the result is to the aesthetic appreciation of normal people.Due to the small amount of data, it is only an experiment. Operating Environment: ubuntu14.04, opencv3.2.0, dlib19.6, python2.7 1. Preparations: 1. Download dlib library and download feature extraction model.

demo for model recognition trained by Tensor Flow 2019-05-01

iterator pattern (less practical) 2019-05-01
1. Introduction to iterator pattern Iterator Pattern, also known as Cursor pattern, is one of the behavioral design patterns. comes from accessing containers, such as list, Map, array, etc. in java. we know that access to container objects must be varied, so we encapsulate traversal methods in containers or do not provide traversal methods. 1. If we encapsulate the traversal method into a container, it will take on too many

Keras 2019-04-30
Keras official website: GitHub: Keras is a highly modularized neural network library, which is implemented in Python and can run on both TensorFlow and Theano.It aims to allow users to carry out the fastest prototype experiment and make the process of turning ideas into results the shortest.Theano and TensorFlow's calculation charts support more general calculations, while Keras specializes in in-depth learning.Theano and TensorFlow are more like in-depth learningNumPy and Keras are Scikit-learn in this field.

Keras Introduction 2019-04-30
1. Introduction to Keras Tool Library 1.1keras IntroductionWhat I'm telling you today is keras. keras is now a very popular tool library, including the fact that tensorflow has incorporated keras into its main code. You can directly tf.keras and call the tool library directly.The reason why keras is mentioned separately is that keras has its unique application scenarios such as laboratories, data competitions and other small environments. Using keras, engineers can spend more time on designing network models instead of coding, and keras is one of the most accessible tool libraries of all.

Keras Learning Note 3: BatchNormalization Layer and Merge Layer 2019-04-30
1. batchNormalization Layer: This layer re-normalizes the activation value of the previous layer on each batch, even if the average value of its output data is close to 0 and its standard deviation is close to 1 keras.layers.normalization.BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001, center=True, scale=True, beta_initializer='zeros', gamma_initializer='ones', moving_mean_initializer='zeros', moving_variance_initializer='ones', beta_regularizer=None, gamma_regularizer=None, beta_constraint=None, gamma_constraint=None) parameters,axis: integer that specifies the axis to normalize, usually the feature axis.For example, axis=1 is generally set after 2D convolution of data_format="

Keras TensorFlow mix has invalid trainable=False setting 2019-04-30
Keras TensorFlow mix has invalid trainable=False setting This is a recent problem, first describe the following problem:First of all, I have a trained model (for example, vgg16). I want to make some changes to this model, for example, add a full connection layer. For various reasons, I can only use TensorFlow to optimize the model. tf optimizer updates the weight of all tf.trainable_variables () by default. The problem lies in