Guest post by Aleksandr Voron To keep OpenVINO Toolkit focused on optimizing and deploying inference, we no longer include OpenCV and DL Streamer in our distribution packages. But not to worry! Both OpenCV and DL Streamer continue to work with OpenVINO and in this blog post we will explain how.  In the 2022.1 release of OpenVINO, OpenCV became an optional dependency that could be installed using a script. This script made it easy to install both OpenVINO and OpenCV together. Starting in the 2022.1.1 release we removed the OpenCV download script and no longer publi ..read more

3D cameras play an important role in the modern industry. They help us solve computer vision tasks that can’t be completed with 2D data. Two years ago, OpenCV supported Orbbec 3D cameras of Astra series. Now here comes more good news. OpenCV supports its 3D UVC cameras! Unlike working with previous Astra 3D cameras which requires OpenCV built with OpenNI2 SDK, no third party libraries are required to work with the new 3D UVC cameras of Astra+ series and Femto series. Plug in the 3D camera and the data from the depth sensor and color sensor or the RGBD streams can be read by using the cv::Video ..read more

Awhile ago we told you how the Computer Vision Annotation Tool (CVAT) has been spun out on its own, thanks in large part to our friends at Intel for years of supporting the project. In this video, we show how to use the relaunched CVAT.ai online tool – how to sign up, upload your data, annotate it, and download it. This video is the first one in a series that’ll guide you on our product and its many features. We hope that after watching it you will be able to use CVAT to its fullest. To get started on your own data annotation projects visit https://www.cvat.ai The post Computer Vision Annotat ..read more

Since the launch last year, the AI model marketplace at Modelplace.AI has been improving by leaps and bounds. The number of models continues to grow along with performance boosts and new features such as model comparison benchmarks. Today we’re excited to tell you about the biggest change yet: Modelplace 2.0. We’ve rethought the entire presentation of the site, thanks to your feedback. Here are some of the biggest new features and how they work: Better For Buyers: The Solutions Catalog Entirely new to the site is the Solutions Catalog, a way to browse not by looking for a specific model, but ..read more

OpenCV AI Scavenger Hunt is a new contest from OpenCV combining the power of the OpenVINO Toolkit with one of the hottest things in tech today: Images generated by feeding text to a speciality trained AI network. Each week in October we’ll email you a AI-generated image and it’s up to you to uncover the prompt we used to generate it. To help you discover the prompt we’re providing you with 2 clues and 2 pretrained models from the OpenVINO Model Zoo. Correctly identifying the prompt will enter you to win a bunch of different cash prizes, including the $3,000 Grand Prize. To participate simply s ..read more

Fast and accurate data annotation is crucial for the success of any AI project.  For the last five years, Intel has developed one of the world’s most popular open-source annotation tools, the Computer Vision Annotation Tool (CVAT), and hosted the tool on CVAT.org. Intel engineers have worked diligently to bring many features to the tool that makes annotation easy for many computer vision tasks like image classification, object detection, and image segmentation. Under the hood of CVAT, AI models, optimized by Intel’s OpenVINO Toolkit, assist these manual tasks to make them faster for human ..read more

Over the last several years OpenCV has been expanding into new areas. This expansion began in earnest with the OpenCV AI Kit product line, and continued with the release of the AI model marketplace at Modelplace.AI. Today we’re happy to announce another big piece of the puzzle: CVAT, the Computer Vision Annotation Tool, is coming home to OpenCV. Many in the OpenCV community know and love CVAT for its ease-of-use, focused user interface, and open source bonafides. What many of you may not know is that CVAT was once a part of OpenCV when both projects were together at Intel. Now, we’re getting ..read more

The contest that began last year is nearing the finish line- with the announcement of our final prize, the winner of the Popular Vote as chosen by you in the OpenCV community, the OpenCV Spatial AI Contest will come to an end. Before we make the big reveal, we at OpenCV would like to take the time to once again thank our sponsors at Intel and Microsoft Azure for making this contest possible, the 50 teams who participated in the finals, and last but not least, to everyone who submitted their projects. If you were chosen or not, it is an achievement to have participated in this momentous contest ..read more