Posted on Leave a comment

Google Cloud AR Computer Vision


google cloud platform []

OAuth 2.0 Scopes for Google APIs []



What’s next for consumer AR in 2020 []

Google Search AR starts rolling out depth-based object blending/occlusion []

The future of augmented reality might be in your ear []



Самый мягкий и пушистый путь в Machine Learning и Deep Neural Networks []

Google shows off stunning new AR features coming to web and mobile apps soon 19
Part of a ARCore’s new Depth API in the works []







Supported Devices []

ARCore develop []

Sceneform Overview []





Augmented Images with ARCore and Sceneform []

Building ARCore apps using Sceneform — Part 1 []

The AR Software Development Leaders of 2018 []

Change scale and proportions of renderable at runtime []

How to change textures in runtime []

How to move object from Anchor to Anchor? []

I want the model to be downloaded from the server []

Is there a way to load models from server []

How to place 3D object at world center []

Dense Pointcloud from depth sensors or stereo []

Introducing the ARCore Depth API for Android and Unity

3D Object Detection []

Extracting Point Cloud []

how to show boundingbox of a node []

Is there a way to draw a Line between two anchors using Sceneform []

How to change renderable color/texture within sceneform? []

Android ArCore Sceneform API. How to change textures in runtime? []

Change Texture of SelectionVisualizer for Selected TransformableNodes []

How ARKit and ARCore recognize vertical planes despite limited depth perception []

12 AR platforms []

8 AR SDK for android and ios []

Best AR SDK for development for iOS and Android in 2018 []

Best available SDK for developing AR applications. []

Best Tools for Building Augmented Reality Mobile Apps []

CraftAR [] []

andresoviedo/android-3D-model-viewer []

virocore []

Building a real-time face detector in Android with ML Kit [] Создание Android приложения для обнаружения лиц в режиме реального времени с использованием Firebase ML Kit []

What I Learned From Interviewing 50 Experts in the AR Industry []

Computer Vision — An Introduction []

Google Identity Platform

OAuth2, Sign-in []

Training id auth by Google []

ML Kit

Mobile Vision

The Mobile Vision API is now a part of ML Kit. We strongly encourage you to try it out, as it comes with new capabilities like on-device image labeling! Also, note that we ultimately plan to wind down the Mobile Vision API, with all new on-device ML capabilities released via ML Kit. Feel free to reach out to Firebase support for help.

Cloud Vision API docs []

Google Cloud Vision API in Android []

Android Example – Programmatically Scan QR Code and Bar Code []

Single/multi view image(s) to voxel reconstruction using a recurrent neural network []

Face detection []

Text recognition []

Visual Positioning Service

Tango is also one of the core technologies behind our new Visual Positioning Service (VPS), which helps devices quickly and accurately understand their location indoors. While GPS is great for getting you to the storefront, with VPS your device can direct you right to the item you’re looking for once inside. VPS works today in partner museums and select Lowe’s stores. We think VPS will be powerful in a variety of scenarios. For instance, imagine how precise location enabled by VPS, combined with audio interfaces, could help visually-impaired people navigate through the world.



Vuforia’s latest object recognition technology


Wikitude Cloud Recognition


Scanning and Detecting 3D Objects (apple)

TensorFlow Lite

Make-up apps

Face recognition / face tracking

Theory and algorithms

      Основная идея распознавания лица состоит в выделении информативных признаков в изображении лица, кодировании этого изображения и сравнении его с информацией, хранящейся в базе данных. В данной работе проводится анализ алгоритмов, базирующихся на методе главных компонент, линейном дискриминантном анализе, обнаружении локальных признаков, с применением вейвлетов Габора, дискретном косинусном преобразовании, локальных бинарных шаблонах. Отмечается, что для корреляционных методов характерна вычислительная сложность и требуются большие объемы памяти, в этой связи на практике целесообразным является применение соответствующих методов, позволяющих уменьшить размерность признаков. Указаны последние разработки компании “Вокорд”, базирующиеся на использовании глубоких нейронных сетей, использующих тестовую базу с миллионом фотографий.
    • метод Виолы-Джонса – первое по сути мощное решение
    • Алгоритм Далала-Триггса
    • by NtechLab based on neural network. One of the best solutions.
      Facial recognition, vehicle identification, text recognition, target tracking

Wall painting

Change hair color

Augmented Images – Video on Wall in ARCore 1.2 Tutorial


80+ лучших инструментов для разработчиков VR и AR [ ]

Demonstrates usage of the Speech SDK to manage voice communication with your Bot-Framework bot, registered with Direct Line Speech channel [ ]

Лучшие бесплатные источники наборов данных для анализа [ ]

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.