wuguangbin1230的博客

The simple task of picking something up is not as easy as it seems. Not for a robot, at least. Roboticists aim to develop a robot that can pick up anything—but today most robots perform “blind graspin

A FT Sensor is a device that is placed between the robot flange (wrist) and the robot tool (end-effector). This device is able to detect force that is applied on the robot tool. For example, if forc

A GelSight sensor attached to a robot’s gripper enables the robot to determine precisely where it has grasped a small screwdriver, removing it from and inserting it back into a slot, even when the gri

引入Discriminative Model的繁荣发展最近，深度学习在很多领域的突破性进展想必不用我多说了。但大家似乎发现了这样的一个现实，即深度学习取得突破性进展的地方貌似都是discriminative的模型。所谓的discriminative可以简单的认为是分类问题，比如给一张图片，判断这张图片上有什么动物；再比如给定一段语音，判断这段语音所对应的文字。在discriminative的模型上...

引言本文主要介绍如何在tensorflow上仅使用200个带标签的mnist图像，实现在一万张测试图片上99%的测试精度，原理在于使用GAN做半监督学习。前文主要介绍一些原理部分，后文详细介绍代码及其实现原理。前文介绍比较简单，有基础的同学请掠过直接看第二部分，文章末尾给出了代码GitHub链接。对GAN不了解的同学可以查看微信公众号：机器学习算法全栈工程师 的GAN入门文章。本博客中的代码最终以...

原论文下载地址：Self-Attention Generative Adversarial Networks该文章作者为：Han ZhangGitHub代码实现：pytorch实现摘要部分：本文提出了Self-Attention Generative Adversarial Networks，在注意力机制，远距离依赖（long-range dependency）下来实现了图片生成任...

1. Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation出自：CVPR 2018,源代码中深度学习框架用的Chainer（1）数据集：用PASCAL VOC 做 source domain用他们自己建的：Clipark1k，Watercolor2k和Comic2k...