如何训练张量流对象检测模型避免检测电视上的人? [英] how to train tensorflow object detection model avoid to detect people on televisons?

问题描述

我使用tensorflow对象检测模型来检测图像上的人，但是图像上还有一些电视，电视上还有人.该模型将检测电视上的人，我如何训练模型以避免检测到电视上的人，谢谢.

I use tensorflow object detection model to detect people on images,but there are　alse some televisions on images, and there are people on television. the model will detect people on television, how can i train the model to avoid to detect people on television, thanks.

推荐答案

大多数电视的目标是呈现逼真的图像.可能无法检测像素代表的是真实图像还是电视图像.

The goal of most television is to present a lifelike image. It's probably impossible to detect if the pixel represents a real image or a TV image.

在很大程度上取决于您的场景，可能有效的是训练一个单独的模型来检测电视.检测到它后，您可以将检测到的区域涂黑并将新图像提供给原始模型.或者，忽略检​​测到的与电视重叠的对象.这将引入误报，导致您无法在某些上下文中检测到某些对象.

What might work, depending a lot on your scenario, is to train a separate model to detect televisions. Once you detect it you can black-out the area detected and feed the new image to your original model. Alternatively, ignore the objects detected that overlap with the television. This will introduce false positives that will cause you to fail to detect certain objects in certain contexts.

这篇关于如何训练张量流对象检测模型避免检测电视上的人?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！