inayatkh
I am a PhD in Computer Vision. Graduated from ICG TU Graz Austria https://scholar.google.com.pk/citations?user=1H3AJbgAAAAJ&hl=en
Pinned Repositories
carla
Open-source simulator for autonomous driving research.
deepFaceDetectRecog
Here, I have demonstrated the use of Multi-task Cascaded Convolutional Neural Networks, [MTCCNN](https://github.com/xiangrufan/keras-mtcnn) for deep face detection and [faceNet](https://arxiv.org/pdf/1503.03832.pdf) for encoding of the aligned faces. Recognition is then performed by matching the faces detected in the test images with the already enrolled or registered faces using euclidean distances.
meetings
erpnext app for learning erpnext developement
practVissionPy3
propms
Includes: Lease, Daily Checklist, Key Set, Meter, Outsourced Attendance. Requires ERPNext. Property Management Solution is powered by [ERPNext](https://github.com/frappe/erpnext), the world's best 100% open source ERP and a comprehensive one system solution that includes accounting, inventory, asset management, HR & Payroll and much more.
realTimeObjectDetection
This is a demonstration of a MobileNetSDD object detector for achieving a very high frame per secons(FPS)
tensorflowCNNfromTrainingToTesting
This is a very Simple example of showing how to build image dataset from your own collection of images, how to train multiple class classifier using tensorflow CNN and how to predict the class of an object in an unseen image. This code is designed so that a newbie user become able to train a tensorflow CNN model using his limited CPU resources. By loading small batches of images in the HDF5 data format is the key for doing so.
tracking-python3
In this repository I will give some implementation of single and multiple object tracking algorithms. These include meanShift, CamShift, Boosting, MIL, KCF, TLD , GoTurn, and MedianFlow. Additionally I will show you how to grab frames at a very high FPS from camera and videos.
ucwinRoadCaffeMobileNet-SSDdetections
Here I have used MobileNet-SSD, deep learning objection technique, for detecting various objects in the UC-Win/Road virutal reality based enviroment. The results of the standard MobileNet-SSD detector was not good enough. Therefore, Transfer Learning Technique is used for fine tuning of the model. Then the new model is applied for detection of the UC-Win/Road objects.
inayatkh's Repositories
inayatkh/tracking-python3
In this repository I will give some implementation of single and multiple object tracking algorithms. These include meanShift, CamShift, Boosting, MIL, KCF, TLD , GoTurn, and MedianFlow. Additionally I will show you how to grab frames at a very high FPS from camera and videos.
inayatkh/tensorflowCNNfromTrainingToTesting
This is a very Simple example of showing how to build image dataset from your own collection of images, how to train multiple class classifier using tensorflow CNN and how to predict the class of an object in an unseen image. This code is designed so that a newbie user become able to train a tensorflow CNN model using his limited CPU resources. By loading small batches of images in the HDF5 data format is the key for doing so.
inayatkh/deepFaceDetectRecog
Here, I have demonstrated the use of Multi-task Cascaded Convolutional Neural Networks, [MTCCNN](https://github.com/xiangrufan/keras-mtcnn) for deep face detection and [faceNet](https://arxiv.org/pdf/1503.03832.pdf) for encoding of the aligned faces. Recognition is then performed by matching the faces detected in the test images with the already enrolled or registered faces using euclidean distances.
inayatkh/ucwinRoadCaffeMobileNet-SSDdetections
Here I have used MobileNet-SSD, deep learning objection technique, for detecting various objects in the UC-Win/Road virutal reality based enviroment. The results of the standard MobileNet-SSD detector was not good enough. Therefore, Transfer Learning Technique is used for fine tuning of the model. Then the new model is applied for detection of the UC-Win/Road objects.
inayatkh/propms
Includes: Lease, Daily Checklist, Key Set, Meter, Outsourced Attendance. Requires ERPNext. Property Management Solution is powered by [ERPNext](https://github.com/frappe/erpnext), the world's best 100% open source ERP and a comprehensive one system solution that includes accounting, inventory, asset management, HR & Payroll and much more.
inayatkh/realTimeObjectDetection
This is a demonstration of a MobileNetSDD object detector for achieving a very high frame per secons(FPS)
inayatkh/carla
Open-source simulator for autonomous driving research.
inayatkh/meetings
erpnext app for learning erpnext developement
inayatkh/caffe
Caffe: a fast open framework for deep learning.
inayatkh/angular_frappe
its modified copy of go1_cms
inayatkh/erpnext
Free and Open Source Enterprise Resource Planning (ERP)
inayatkh/frappe
Low code web framework for real world applications, in Python and Javascript
inayatkh/health
Open Source Health Information System
inayatkh/hotelinventoryApp
inayatkh/hrms
Open Source HR and Payroll Software
inayatkh/Labelbox
Labelbox is the fastest way to annotate data to build and ship artificial intelligence applications.
inayatkh/MDXeMotionV2
Python code for second generation motion platform
inayatkh/ml-agents
Unity Machine Learning Agents Toolkit
inayatkh/models
Models and examples built with TensorFlow
inayatkh/payments
A payments app for frappe
inayatkh/personCounterNMob
inayatkh/portfolio-sass
A modern portfolio using Sass.
inayatkh/PythonRobotics
Python sample codes for robotics algorithms.
inayatkh/ReactSocialApp
This is an example project for creating a simple social app with React.js
inayatkh/stewart
Simulating a Stewart platform in Gazebo using a plugin to allow control of a closed loop manipulator with ROS.
inayatkh/tensorflow
Computation using data flow graphs for scalable machine learning
inayatkh/test
dsfasdf
inayatkh/testhtml
just testing html documentation browsing
inayatkh/WashoutFilter
Washout filter processes a vehicle motion to produce a simulator motion.
inayatkh/Windows-universal-samples
API samples for the Universal Windows Platform.