人工智能Java SDK:实例分割(支持80分类)

实例分割(支持80分类)SDK

支持对图片中80个分类的目标进行实例分割。

支持coco数据集分类如下:

  • person
  • bicycle
  • car
  • motorcycle
  • airplane
  • bus
  • train
  • truck
  • boat
  • traffic light
  • fire hydrant
  • stop sign
  • parking meter
  • bench
  • bird

点击下载

SDK算法:

  • InstanceSegmentation
    mask_rcnn, backbone: resnet18,dataset: coco

运行例子 - InstanceSegmentationExample

  • 测试图片
    人工智能Java SDK:实例分割(支持80分类)

运行成功后,命令行应该看到下面的信息:

...
[INFO ] - [
	class: "dog", probability: 0.99422, bounds: [x=0.024, y=0.344, width=0.225, height=0.542]
	class: "person", probability: 0.97572, bounds: [x=0.263, y=0.139, width=0.209, height=0.474]
	class: "person", probability: 0.97018, bounds: [x=0.236, y=0.148, width=0.547, height=0.685]
	class: "cat", probability: 0.88282, bounds: [x=0.276, y=0.551, width=0.249, height=0.194]
	class: "chair", probability: 0.55487, bounds: [x=0.716, y=0.249, width=0.187, height=0.276]
]

帮助

引擎定制化配置,可以提升首次运行的引擎下载速度,解决外网无法访问或者带宽过低的问题。
引擎定制化配置

官网:

http://www.aias.top/

Github地址:

https://github.com/mymagicpower/AIAS
https://gitee.com/mymagicpower/AIAS

上一篇:1.qs100模块简单使用笔记(电信物联网平台ctwing)


下一篇:使用 node 调用支付宝支付接口