zfxu Goto Github PK
Name: Zhifei Xu
Type: User
Bio: Computer Vision Engineer
Location: Chengdu
Name: Zhifei Xu
Type: User
Bio: Computer Vision Engineer
Location: Chengdu
基于tensorflow的hourglass模型demo
Official PyTorch Implementation of Hypercorrelation Squeeze for Few-Shot Segmentation, arXiv 2021
2019 Huawei Cloud garbage Classification Competition online score 2th place
This repo contains the eval code for Hybrid-E-loss, which is written by PyTorch code.
A MATLAB pipeline to create hybrid 3D image data with real point clouds and CG models
基于深度学习高性能中文车牌识别 High Performance Chinese License Plate Recognition Framework.
ICCV 2021 论文和开源项目合集
Efficient 2×2 block-based connected components labeling algorithms
ICNet for Real-Time Semantic Segmentation on High-Resolution Images
An ICP library with Matlab bindings
Non local means filter for ImageJ
Detect ridges / lines with imagej
Intensity-based_Lidar_Camera_Calibration
Image alignment using least squares NCC
Open source book on industrial machine vision techniques.
Object Detection in general means machines method to locate and label objects. These techniques can be used for both static and dynamic data. These techniques are already in wide use in many industries such as smart phones, sports, smart vehicles. Various methods have been adopted to get the optimal way for recognizing the objects in an image and many of these techniques have a major problem in space management and communication between multiple different platforms. Open source libraries such as OpenCV’s DNN library and tensorflow Object detection API offer easy-to-use, open source frameworks where pre trained models for object detection reaches high accuracy. However the main aim being the space management and versatility of the model among different platforms is the main focus of study here. We have developed an object recognition model using convolutional neural networks that is trained and tested with the CIFAR-10 dataset. The model is based on linear regression as we are using the same to train and test our model. Also since the training process is complex, time consuming and space consuming, we will store our training data in a .h5 file. Then we can use the same file for the same model or a different model using the same training data as a part of multiple models used in the detector. In this project TensorFlow and Keras API are used to facilitate the process of building, training and testing the model. Tensor flow is a high level library for numerical computation. It helps us to build machine learning model, however we have to build each layer within our network and manually build training and testing loops and optimizers. It is Flexible and easy to use but keras acts as an API that works on TensorFlow to make all the attributes of a model easily accessible. Keras basically works on top of the other machine learning frameworks. While building our model we will see the difference in accuracy that we will get when we are using different activation functions (such as Tanh, Relu, LeakyRelu, Softmax ) for the same model. The Object recognizer that we are going to built can identify objects from ten different domains that are mentioned in the CIFAR-10 dataset. This dataset contains 60,000 32x32 images, 6000 images in each class. There are 50,000 training images and 10,000 testing images in random order.This model can act as a base for a more complex object recognizer, It can be used as a part of a larger model consisting a large no of models. This model can be used for quick load applications and mobile applications as well. The programming for the following project has been done using python ver.3.5.1. Also the other in built libraries have been used.
TensorBasicModel
Implementation of Segnet, FCN, UNet and other models in Keras.
Implementation of Super Resolution CNN in Keras.
A general list of resources to image text localization and recognition 场景文本位置感知与识别的论文资源汇总
Load Image Dataset with OpenCV-Python to NumpyArray forTensorFlow or Keras
DenseFuse (IEEE TIP 2019) - TensorFlow 1.8.0
ImageNet model implemented using the Keras Functional API
「画像処理100本ノック」中文版本!为图像处理初学者设计的 100 个问题。
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.