Unity物体和场景识别
CoderZ1010 人气:0接口介绍:
可识别超过10万类常见物体和场景,接口返回大类及细分类的名称,并支持获取识别结果对应的百科信息;还可使用EasyDL定制训练平台,定制识别分类标签。广泛适用于图像或视频内容分析、拍照识图等业务场景。
创建应用:
在产品服务中搜索图像识别,创建应用,获取AppID、APIKey、SecretKey信息:
查阅官方文档,以下是通用物体和场景识别接口返回数据参数详情:
定义数据结构:
using System; /// <summary> /// 通用物体和场景识别 /// </summary> [Serializable] public class AdvancedGeneralRecognition { /// <summary> /// 唯一的log id,用于问题定位 /// </summary> public float log_id; /// <summary> /// 返回结果数目,及result数组中的元素个数 /// </summary> public int result_num; /// <summary> /// 返回结果数组 /// </summary> public AdvancedGeneralRecognitionResult[] result; } /// <summary> /// 通用物体和场景识别结果 /// </summary> [Serializable] public class AdvancedGeneralRecognitionResult { /// <summary> /// 图片中的物体或场景名称 /// </summary> public string keyword; /// <summary> /// 置信度,0-1 /// </summary> public float score; /// <summary> /// 识别结果的上层标签,有部分钱币、动漫、烟酒等tag无上层标签 /// </summary> public string root; /// <summary> /// 百科词条 /// </summary> public BaikeInfo baike_info; } [Serializable] public class BaikeInfo { /// <summary> /// 对应识别结果百度百科页面链接 /// </summary> public string baike_url; /// <summary> /// 对应识别结果百科图片链接 /// </summary> public string image_url; /// <summary> /// 对应识别结果百科内容描述 /// </summary> public string description; }
下载C# SDK:
下载完成后将AipSdk.dll动态库导入到Unity中:
以下是调用接口时传入的参数详情:
封装调用函数:
using System; using System.Collections.Generic; using UnityEngine; /// <summary> /// 图像识别 /// </summary> public class ImageRecognition { //以下信息于百度开发者中心控制台创建应用获取 private const string appID = ""; private const string apiKey = ""; private const string secretKey = ""; /// <summary> /// 通用物体和场景识别 /// </summary> /// <param name="bytes">图片字节数据</param> /// <param name="baikeNum">返回百科信息的结果数,默认不返回</param> /// <returns></returns> public static AdvancedGeneralRecognition AdvancedGeneral(byte[] bytes, int baikeNum = 0) { var client = new Baidu.Aip.ImageClassify.ImageClassify(apiKey, secretKey); try { var options = new Dictionary<string, object> { { "baike_num", baikeNum }, }; var response = client.AdvancedGeneral(bytes, options); AdvancedGeneralRecognition recognition = JsonConvert.DeserializeObject<AdvancedGeneralRecognition>(response.ToString()); return recognition; } catch (Exception error) { Debug.LogError(error); } return null; } }
测试图片:
using System.IO; using UnityEngine; public class Example : MonoBehaviour { private void Start() { ImageRecognition.AdvancedGeneral(File.ReadAllBytes(Application.dataPath + "/Picture.jpg"), 1); } }
加载全部内容