VRT Visual Recognition Tool

VRT makes it easy to add image analysis to your applications. You provide an image to the API and the service detects objects, people, faces; extracts text (lines and words with geometry), scenes and activities. It provides content moderation API to detect any inappropriate, unwanted, or offensive content. It can also detect celebrities (with their emotions), labels, personal protective equipment (PPE) and provide highly accurate facial analysis (bounding box, landmarks, age range, gender, ...

搜集汇总

背景与挑战

背景概述

VRT视觉识别工具提供图像分析API，能够检测图像中的对象、人脸、文本、场景和活动，并支持内容审核功能。此外，它还能识别名人、标签、个人防护装备，并执行高精度的面部分析，包括边界框、关键点、年龄范围和性别等属性。