这些计算机视觉 API 可以检测多少对象?
How many objects these Computer Vision API can detect?
https://docs.microsoft.com/fr-fr/azure/cognitive-services/computer-vision/concept-object-detection
https://cloud.google.com/vision/docs/object-localizer
我想知道使用论文 API 可以识别多少对象和哪些对象,但我找不到关于该事实的提及。
我发现 google API 使用基于 schema.org 类型的 https://developers.google.com/knowledge-graph/ 但我不太了解它的全部内容。
抱歉,据我所知,Azure Computer Vision 无法检测到 类 的固定列表。
顺便说一下,即使有一个,这个列表也会定期更新(但没有公布时间表)。
无论如何,都有局限性(见文档here):
It's important to note the limitations of object detection so you can
avoid or mitigate the effects of false negatives (missed objects) and
limited detail.
- Objects are generally not detected if they're small (less than 5% of
the image).
- Objects are generally not detected if they're arranged
closely together (a stack of plates, for example).
- Objects are not differentiated by brand or product names (different
types of sodas on a store shelf, for example). However, you can get
brand information from an image by using the Brand detection feature.
如果您想检测特定对象,我强烈建议使用 Custom Vision
(文档/概述 here),而不是 Computer Vision
,您可以在其中使用自己的模型进行训练与您要检测的内容相匹配的图像
https://docs.microsoft.com/fr-fr/azure/cognitive-services/computer-vision/concept-object-detection
https://cloud.google.com/vision/docs/object-localizer
我想知道使用论文 API 可以识别多少对象和哪些对象,但我找不到关于该事实的提及。
我发现 google API 使用基于 schema.org 类型的 https://developers.google.com/knowledge-graph/ 但我不太了解它的全部内容。
抱歉,据我所知,Azure Computer Vision 无法检测到 类 的固定列表。
顺便说一下,即使有一个,这个列表也会定期更新(但没有公布时间表)。
无论如何,都有局限性(见文档here):
It's important to note the limitations of object detection so you can avoid or mitigate the effects of false negatives (missed objects) and limited detail.
- Objects are generally not detected if they're small (less than 5% of the image).
- Objects are generally not detected if they're arranged closely together (a stack of plates, for example).
- Objects are not differentiated by brand or product names (different types of sodas on a store shelf, for example). However, you can get brand information from an image by using the Brand detection feature.
如果您想检测特定对象,我强烈建议使用 Custom Vision
(文档/概述 here),而不是 Computer Vision
,您可以在其中使用自己的模型进行训练与您要检测的内容相匹配的图像