Computer Vision Opplæringskurs

Caffe er en dyp læringsramme laget med uttrykk, hastighet og modularitet i tankene.

Dette kurset utforsker anvendelsen av Caffe som en Deep læringsramme for bildegjenkjenning ved bruk av MNIST som eksempel

Publikum

Dette kurset passer for Deep Learning forskere og ingeniører som er interessert i å bruke Caffe som rammeverk.

Etter fullført kurs vil delegatene kunne:

- forstå Caffe sin struktur og distribusjonsmekanismer
- utføre installasjons- / produksjonsmiljø / arkitekturoppgaver og konfigurasjon
- vurdere kodekvalitet, utfør feilsøking, overvåking
- implementere avansert produksjon som treningsmodeller, implementere lag og logging

Marvin is an extensible, cross-platform, open-source image and video processing framework developed in Java. Developers can use Marvin to manipulate images, extract features from images for classification tasks, generate figures algorithmically, process video file datasets, and set up unit test automation.

Some of Marvin's video applications include filtering, augmented reality, object tracking and motion detection.

In this instructor-led, live course participants will learn the principles of image and video analysis and utilize the Marvin Framework and its image processing algorithms to construct their own application.

Format of the Course

- The basic principles of image analysis, video analysis and the Marvin Framework are first introduced. Students are given project-based tasks which allow them to practice the concepts learned. By the end of the class, participants will have developed their own application using the Marvin Framework and libraries.

OpenCV (Open Source Computer Vision Library: http://opencv.org) er et åpen kildekode-BSD-lisensiert bibliotek som inkluderer flere hundre datamaskinvisjonsalgoritmer.

Publikum

Dette kurset er rettet mot ingeniører og arkitekter som søker å benytte seg av OpenCV til prosjekter med datasyn

OpenCV is a library of programming functions for deciphering images with computer algorithms. OpenCV 4 is the latest OpenCV release and it provides optimized modularity, updated algorithms, and more. With OpenCV 4 and Python, users will be able to view, load, and classify images and videos for advanced image recognition.

This instructor-led, live training (online or onsite) is aimed at software engineers who wish to program in Python with OpenCV 4 for deep learning.

By the end of this training, participants will be able to:

- View, load, and classify images and videos using OpenCV 4.
- Implement deep learning in OpenCV 4 with TensorFlow and Keras.
- Run deep learning models and generate impactful reports from images and videos.

Format of the Course

- Interactive lecture and discussion.
- Lots of exercises and practice.
- Hands-on implementation in a live-lab environment.

Course Customization Options

- To request a customized training for this course, please contact us to arrange.

Pattern Matching er en teknikk som brukes til å finne spesifiserte mønstre i et bilde. Det kan brukes til å bestemme eksistensen av spesifiserte egenskaper i et tatt bilde, for eksempel den forventede etiketten på et mangelfullt produkt på en fabrikklinje eller de spesifiserte dimensjonene til en komponent. Det er forskjellig fra " Pattern Recognition " (som gjenkjenner generelle mønstre basert på større samlinger av relaterte prøver) ved at den spesifikt dikterer hva vi leter etter, for så å fortelle oss om det forventede mønsteret eksisterer eller ikke.

Kursets format

- Dette kurset introduserer tilnærminger, teknologier og algoritmer som brukes i feltet mønster matching som det gjelder Machine Vision .

Computer Vision er et felt som innebærer automatisk å trekke ut, analysere og forstå nyttig informasjon fra digitale medier. Python er et programmeringsspråk på høyt nivå kjent for sin klare syntaks og kodelesbarhet.

I denne instruktørledede, liveopplæringen, vil deltakerne lære det grunnleggende om Computer Vision når de går gjennom opprettelsen av et sett med enkel Computer Vision-applikasjon ved hjelp av Python .

Ved slutten av denne opplæringen vil deltakerne kunne:

- Forstå det grunnleggende i Computer Vision
- Bruk Python til å implementere Computer Vision-oppgaver
- Bygg sine egne ansikts-, objekt- og bevegelsesdeteksjonssystemer

Publikum

- Python programmerere som er interessert i Computer Vision

Kursets format

- Delforelesning, deldiskusjon, øvelser og tung praktisk øvelse

This instructor-led, live training introduces the software, hardware, and step-by-step process needed to build a facial recognition system from scratch. Facial Recognition is also known as Face Recognition.

The hardware used in this lab includes Rasberry Pi, a camera module, servos (optional), etc. Participants are responsible for purchasing these components themselves. The software used includes OpenCV, Linux, Python, etc.

By the end of this training, participants will be able to:

- Install Linux, OpenCV and other software utilities and libraries on a Rasberry Pi.
- Configure OpenCV to capture and detect facial images.
- Understand the various options for packaging a Rasberry Pi system for use in real-world environments.
- Adapt the system for a variety of use cases, including surveillance, identity verification, etc.

Format of the course

- Part lecture, part discussion, exercises and heavy hands-on practice

Note

- Other hardware and software options include: Arduino, OpenFace, Windows, etc. If you wish to use any of these, please contact us to arrange.

Deep learning is a subfield of machine learning. It uses methods based on learning data representations and structures such as neural networks.

Keras is a high-level neural networks API for fast development and experimentation. It runs on top of TensorFlow, CNTK, or Theano.

This instructor-led, live training (online or onsite) is aimed at developers who wish to build a self-driving car (autonomous vehicle) using deep learning techniques.

By the end of this training, participants will be able to:

- Use computer vision techniques to identify lanes.
- Use Keras to build and train convolutional neural networks.
- Train a deep learning model to differentiate traffic signs.
- Simulate a fully autonomous car.

Format of the Course

- Interactive lecture and discussion.
- Lots of exercises and practice.
- Hands-on implementation in a live-lab environment.

Course Customization Options

- To request a customized training for this course, please contact us to arrange.

SimpleCV er et rammeverk med åpen kildekode - noe som betyr at det er en samling biblioteker og programvare som du kan bruke til å utvikle visjonsapplikasjoner. Det lar deg jobbe med bilder eller videostrømmer som kommer fra webkameraer, Kinects, FireWire og IP-kameraer, eller mobiltelefoner. Det hjelper deg med å bygge programvare for å gjøre de forskjellige teknologiene dine ikke bare til å se verden, men også forstå den.

Publikum

Dette kurset er rettet mot ingeniører og utviklere som søker å utvikle datavisjonsapplikasjoner med SimpleCV.

Video analytics refers to the technology and techniques used to process a video stream. A common application would be capturing and identifying live video events through motion detection, facial recognition, crowd and vehicle counting, etc.

This instructor-led, live training (online or onsite) is aimed at developers who wish to build hardware-accelerated object detection and tracking models to analyze streaming video data.

By the end of this training, participants will be able to:

- Install and configure the necessary development environment, software and libraries to begin developing.
- Build, train, and deploy deep learning models to analyze live video feeds.
- Identify, track, segment and predict different objects within video frames.
- Optimize object detection and tracking models.
- Deploy an intelligent video analytics (IVA) application.

Format of the Course

- Interactive lecture and discussion.
- Lots of exercises and practice.
- Hands-on implementation in a live-lab environment.

Course Customization Options

- To request a customized training for this course, please contact us to arrange.

YOLO (You Only Look Once) is an algorithm turned into pre-trained models for object detection. It is tested by the Darknet neural network framework, making it ideal for developing computer vision features based on the COCO (Common Objects in Context) dataset. The latest variants of the YOLO framework, YOLOv3-v4, allows programs to efficiently execute object locating and classifying tasks while running in real-time.

This instructor-led, live training (online or onsite) is aimed at backend developers and data scientists who wish to incorporate pre-trained YOLO models into their enterprise-driven programs and implement cost-effective components for object-detection.

By the end of this training, participants will be able to:

- Install and configure the necessary tools and libraries required in object detection using YOLO.
- Customize Python command-line applications that operate based on YOLO pre-trained models.
- Implement the framework of pre-trained YOLO models for various computer vision projects.
- Convert existing datasets for object detection into YOLO format.
- Understand the fundamental concepts of the YOLO algorithm for computer vision and/or deep learning.

Format of the Course

- Interactive lecture and discussion.
- Lots of exercises and practice.
- Hands-on implementation in a live-lab environment.

Course Customization Options

- To request a customized training for this course, please contact us to arrange.