Blogs by Ashwin

  Overview Producing material in video format is becoming more and more prevalent for both individuals and companies. For instance, you might record a teleconference that includes webcam footage and slide or document presentations, or you might use a cellphone to record a live event. Because of this, video files include a lot of information that may need to be extracted for analysis or to facilitate indexing for searchability. Azure Video Indexer is a service that extracts information from videos, such as scene segmentation, item labels, word recognition, face identification, and more. Understand Azure Video Indexer Capabilities The Azure Video Indexer service is designed to help you extract information from videos. It provides functionality that you can use for: Facial recognition - detecting the presence of individual people in the image. This requires Access approval. Optical character recognition - reading text in the video. Speech transcription - creating a text transcript ...

  Introduction Azure's AI Vision service uses algorithms to process images and return information. Let's say you are tasked with transferring the text from thousands of photos to a computer database. The text in the scanned photos is multilingual and arranged in several formats. How do you ensure that the data is input accurately while finishing the project in a fair amount of time? Every day, businesses all across the world deal with similar situations. The project would be difficult to finish without AI services, particularly if its scope changed. We can use Optical Character Recognition (OCR) and handle this project as an Azure AI Vision scenario using AI services. OCR enables you to extract text from documents, including handwritten or unstructured papers, as well as from photographs, such as pictures of products and street signs. Machine learning models must be trained to address a wide range of use cases in order to create an automated AI solution.  The Azure AI Vision ...

  Understand Capabilities Of The Face Service The Face service provides comprehensive facial detection, analysis, and recognition capabilities. The Face service provides functionality that you can use for: Face detection - for each detected face, the results include an ID that identifies the face and the bounding box coordinates indicating its location in the image. Face attribute analysis - you can return a wide range of facial attributes, including- Head pose (pitch, roll, and yaw orientation in 3D space) Glasses (NoGlasses, ReadingGlasses, Sunglasses, or Swimming Goggles) Blur (low, medium, or high) Exposure (underExposure, goodExposure, or overExposure) Noise (visual noise in the image) Occlusion (objects obscuring the face) Accessories (glasses, headwear, mask) QualityForRecognition (low, medium, or high) Facial landmark location - coordinates for key landmarks in relation to facial features (for example, eye corners, pupils, tip of nose, and so on). Face comparison - you ...

  Introduction AI systems frequently face computer vision difficulties like face detection, analysis, and recognition. One of the main ways AI systems can behave like humans and develop empathy with users is by being able to recognize a person based on their facial traits, determine whether a person is present, or identify a person's facial position. Identify Options For Face Detection Analysis and Identification There are two Azure AI services that you can use to build solutions that detect faces or people in images. The Azure AI Vision Service The Azure AI Vision service enables you to detect people in an image, as well as returning a bounding box for its location. The Face Service The Face service offers more comprehensive facial analysis capabilities than the Azure AI Vision service, including: Face detection (with bounding box). Comprehensive facial feature analysis (including head pose, presence of spectacles, blur, facial landmarks, occlusion and others). Face comparison and...

  Train an Object Detector You can either utilize the REST API or SDK to build code that carries out the training activities, or you can use the Azure AI Custom Vision portal to upload and label photos prior to training, assessing, testing, and releasing the model. The labeling of the images with tags is the main distinction between training an object detection model and training an image classification model. Object detection necessitates that each label have a tag and a region that specifies the bounding box for each object in an image, whereas image classification requires one or more tags that apply to the entire image. You can label your training photos using the graphical interface offered by the Azure AI Custom Vision portal. Consider Options for Labeling Images The Azure AI Custom Vision portal's interactive interface is the simplest way to identify photos for object detection. By sliding the bounding box to enclose the object you wish to label, you can modify or add tags t...

  Introduction To find and identify items in pictures, object detection is utilized. Azure AI Custom Vision can be used to train a model to identify particular object classes in pictures. A common computer vision challenge is object detection, which calls for software to locate particular object classes in an image. Understand Object Detection A model is trained to identify the existence and placement of one or more kinds of objects in an image in object detection, a type of computer vision. For example, a grocery store's AI-enabled checkout system could have to determine the kind and location of goods a consumer is buying. There are two components to an object detection prediction: The class label of each object detected in the image. For example, you might ascertain that an image contains one apple and two oranges. The location of each object within the image, indicated as coordinates of a bounding box that encloses the object. Use the Azure AI Custom Vision Service For Object De...