Programmer Think

Realization idea Pixelation is to make many adjacent pixels become one pixel, which will naturally have the effect of pixelation. According to this idea, we can immediately think of a way to divide the screen into several areas. Each area is composed of several adjacent pixels. Each area is a square, and then calculate the convolution of the s ...

Deep learning based on Keras (II) -- construction and training of LeNet LeNet is a very efficient convolutional neural network for handwritten character recognition. Although the network is small, it contains the basic modules of deep learning: convolution layer, pooling layer and full connection layer. It is also the basis of other deep l ...

preface During this period of time, I have benefited a lot from watching the explanation of in-depth learning by up Master Liu Er of station b. Let me know what Inception is and how to embed the structure into the model. Here is the video address: https://www.bilibili.com/video/BV1Y7411d7Ys?p=11&share_source=copy_web Interested fri ...

Summary of Classic Network Structure--MobileNet Series MobileNet mobilenet was proposed by Google. Advantages: Small size, small computation, suitable for convolution neural network of mobile devices. Classification/Target Detection/Semantic Segmentation can be implemented; Miniaturization: The convolution kernel is decomposed and replaced ...

Post processing principle and process Post processing is to process the image after rendering, and unity uses C# script It is not displayed after rendering, take one more step, and then display it on the screen after script processing, so as to realize various blur effects, Bloom, stroke and so on General process of post-treatment Obt ...