diff --git a/.gitignore b/.gitignore
index e377bf6..35b492e 100644
--- a/.gitignore
+++ b/.gitignore
@@ -15,6 +15,9 @@ share/python-wheels/
 *.egg
 MANIFEST
 .idea/
+models/
+models-converting.sh
+models-downloading.sh
 
 # Prerequisites
 *.d
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 44707a4..065f970 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -271,6 +271,10 @@ add_executable(GimbalLandingMarkerDetection samples/demo/gimbal_landing_marker_d
 target_link_libraries(GimbalLandingMarkerDetection sv_world)
 add_executable(GimbalUdpDetectionInfoSender samples/demo/gimbal_udp_detection_info_sender.cpp)
 target_link_libraries(GimbalUdpDetectionInfoSender sv_world)
+add_executable(ArucoDetectionWithSingleObjectTracking samples/demo/aruco_detection_with_single_object_tracking.cpp)
+target_link_libraries(ArucoDetectionWithSingleObjectTracking sv_world)
+add_executable(CarDetectionWithTracking samples/demo/car_detection_with_tracking.cpp)
+target_link_libraries(CarDetectionWithTracking sv_world)
 
 add_executable(EvalFpsOnVideo samples/test/eval_fps_on_video.cpp)
 target_link_libraries(EvalFpsOnVideo sv_world)
diff --git a/calib_webcam_1280x720.yaml b/calib_webcam_1280x720.yaml
new file mode 100644
index 0000000..39df5ce
--- /dev/null
+++ b/calib_webcam_1280x720.yaml
@@ -0,0 +1,20 @@
+%YAML:1.0
+---
+calibration_time: "2021年01月12日 星期二 18时08分01秒"
+image_width: 1280
+image_height: 720
+flags: 0
+camera_matrix: !!opencv-matrix
+   rows: 3
+   cols: 3
+   dt: d
+   data: [ 7.9379415710551370e+02, 0., 2.9783879354295328e+02, 0.,
+       7.9491985564466654e+02, 3.0942416136837386e+02, 0., 0., 1. ]
+distortion_coefficients: !!opencv-matrix
+   rows: 1
+   cols: 5
+   dt: d
+   data: [ 2.0950200339181715e-01, -1.1587468096518483e+00,
+       5.5342063671841328e-03, 2.2214393775334758e-04,
+       1.7127431916651392e+00 ]
+avg_reprojection_error: 2.8342964851391211e-01
diff --git a/calib_webcam_640x480.yaml b/calib_webcam_640x480.yaml
new file mode 100644
index 0000000..54f1526
--- /dev/null
+++ b/calib_webcam_640x480.yaml
@@ -0,0 +1,20 @@
+%YAML:1.0
+---
+calibration_time: "2023年07月14日 星期五 16时39分17秒"
+image_width: 640
+image_height: 480
+flags: 0
+camera_matrix: !!opencv-matrix
+   rows: 3
+   cols: 3
+   dt: d
+   data: [ 4.5099311307542973e+02, 0., 3.2898947972890943e+02, 0.,
+       6.0215873600107579e+02, 2.4195307609106428e+02, 0., 0., 1. ]
+distortion_coefficients: !!opencv-matrix
+   rows: 1
+   cols: 5
+   dt: d
+   data: [ 1.0737258446369682e-01, -1.2782122264046064e-01,
+       1.6844258609297487e-03, -6.6256775118868144e-04,
+       -3.5333889479158398e-01 ]
+avg_reprojection_error: 3.3968000452388564e-01
diff --git a/samples/calib/calibrate_camera_charuco.cpp b/samples/calib/calibrate_camera_charuco.cpp
index 7337dd2..d6effe3 100644
--- a/samples/calib/calibrate_camera_charuco.cpp
+++ b/samples/calib/calibrate_camera_charuco.cpp
@@ -119,6 +119,8 @@ int main(int argc, char *argv[]) {
     }
 
     VideoCapture inputVideo;
+   
+   
     int waitTime;
     if(!video.empty()) {
         inputVideo.open(video);
@@ -127,6 +129,9 @@ int main(int argc, char *argv[]) {
         inputVideo.open(camId);
         waitTime = 10;
     }
+    
+  
+    
 
     aruco::Dictionary dictionary = aruco::getPredefinedDictionary(0);
     if (parser.has("d")) {
@@ -159,6 +164,8 @@ int main(int argc, char *argv[]) {
     while(inputVideo.grab()) {
         Mat image, imageCopy;
         inputVideo.retrieve(image);
+        cv::resize(image, image, cv::Size(640, 480));
+       
 
         vector< int > ids;
         vector< vector< Point2f > > corners, rejected;
diff --git a/samples/demo/aruco_detection.cpp b/samples/demo/aruco_detection.cpp
index 0229400..83fbabb 100644
--- a/samples/demo/aruco_detection.cpp
+++ b/samples/demo/aruco_detection.cpp
@@ -9,13 +9,13 @@ int main(int argc, char *argv[]) {
   // 实例化Aruco检测器类
   sv::ArucoDetector ad;
   // 手动导入相机参数，如果使用Amov的G1等吊舱或相机，则可以忽略该步骤，将自动下载相机参数文件
-  ad.loadCameraParams(sv::get_home() + "/SpireCV/calib_webcam_640x480.yaml");
+  ad.loadCameraParams(sv::get_home() + "/SpireCV/calib_webcam_1280x720.yaml");
   
   // 打开摄像头
   sv::Camera cap;
   // cap.setWH(640, 480);
   // cap.setFps(30);
-  cap.open(sv::CameraType::WEBCAM, 0);  // CameraID 0
+  cap.open(sv::CameraType::WEBCAM, 2);  // CameraID 0
   // 实例化OpenCV的Mat类，用于内存单帧图像
   cv::Mat img;
   int frame_id = 0;
diff --git a/samples/demo/aruco_detection_with_single_object_tracking.cpp b/samples/demo/aruco_detection_with_single_object_tracking.cpp
new file mode 100644
index 0000000..356d55b
--- /dev/null
+++ b/samples/demo/aruco_detection_with_single_object_tracking.cpp
@@ -0,0 +1,173 @@
+
+#include <iostream>
+#include <string>
+// 包含SpireCV SDK头文件
+#include <sv_world.h>
+
+using namespace std;
+
+// 定义窗口名称
+static const std::string RGB_WINDOW = "Image window";
+// 框选到的矩形
+cv::Rect rect_sel;
+// 框选起始点
+cv::Point pt_origin;
+// 是否按下左键
+bool b_clicked = false;
+// 是否得到一个新的框选区域
+bool b_renew_ROI = false;
+// 是否开始跟踪
+bool b_begin_TRACK = false;
+// 实现框选逻辑的回调函数
+void onMouse(int event, int x, int y, int, void*);
+
+int main(int argc, char *argv[]) {
+  // 定义一个新的窗口，可在上面进行框选操作
+  cv::namedWindow(RGB_WINDOW);
+  // 设置窗口操作回调函数，该函数实现整个框选逻辑
+  cv::setMouseCallback(RGB_WINDOW, onMouse, 0);
+  // 实例化 框选目标跟踪类
+  sv::SingleObjectTracker sot;
+  // 手动导入相机参数，如果使用Amov的G1等吊舱或相机，则可以忽略该步骤，将自动下载相机参数文件
+  sot.loadCameraParams(sv::get_home() + "/SpireCV/calib_webcam_640x480.yaml");
+  // sot.loadCameraParams(sv::get_home() + "/SpireCV/calib_webcam_1280x720.yaml");
+  // sot.loadCameraParams(sv::get_home() + "/SpireCV/calib_webcam_1920x1080.yaml");
+  
+  // 实例化Aruco检测器类
+  sv::ArucoDetector ad;
+  // 手动导入相机参数，如果使用Amov的G1等吊舱或相机，则可以忽略该步骤，将自动下载相机参数文件
+  ad.loadCameraParams(sv::get_home() + "/SpireCV/calib_webcam_640x480.yaml");
+
+  // 打开摄像头
+  sv::Camera cap;
+  cap.setWH(640, 480);
+  cap.setFps(30);
+  cap.open(sv::CameraType::WEBCAM, 2);  // CameraID 0
+  // cv::VideoCapture cap("/home/amov/SpireCV/test/tracking_1280x720.mp4");
+  // 实例化OpenCV的Mat类，用于内存单帧图像
+  cv::Mat img;
+  int frame_id = 0;
+  while (1)
+  {
+    // 实例化SpireCV的 单帧检测结果 接口类 TargetsInFrame
+    sv::TargetsInFrame tgts(frame_id++);
+    // 读取一帧图像到img
+    cap.read(img);
+    cv::resize(img, img, cv::Size(sot.image_width, sot.image_height));
+
+    // 开始 单目标跟踪 逻辑
+    // 是否有新的目标被手动框选
+    if (b_renew_ROI)
+    {
+      // 拿新的框选区域 来 初始化跟踪器
+      sot.init(img, rect_sel);
+      // std::cout << rect_sel << std::endl;
+      // 重置框选标志
+      b_renew_ROI = false;
+      // 开始跟踪
+      b_begin_TRACK = true;
+    }
+    else if (b_begin_TRACK)
+    {
+      // 以前一帧的结果继续跟踪
+      sot.track(img, tgts);
+
+      // 执行Aruco二维码检测
+      ad.detect(img, tgts);
+
+      // 可视化检测结果，叠加到img上
+      sv::drawTargetsInFrame(img, tgts);
+
+      // 控制台打印 单目标跟踪 结果
+      printf("Frame-[%d]\n", frame_id);
+      // 打印当前检测的FPS
+      printf("  FPS = %.2f\n", tgts.fps);
+      // 打印当前相机的视场角（degree）
+      printf("  FOV (fx, fy) = (%.2f, %.2f)\n", tgts.fov_x, tgts.fov_y);
+      // 打印当前输入图像的像素宽度和高度
+      printf("  Frame Size (width, height) = (%d, %d)\n", tgts.width, tgts.height);
+      if (tgts.targets.size() > 0)
+      {
+        printf("Frame-[%d]\n", frame_id);
+        // 打印 跟踪目标 的中心位置，cx，cy的值域为[0, 1]，以及cx，cy的像素值
+        printf("  Tracking Center (cx, cy) = (%.3f, %.3f), in Pixels = ((%d, %d))\n",
+          tgts.targets[0].cx, tgts.targets[0].cy,
+          int(tgts.targets[0].cx * tgts.width),
+          int(tgts.targets[0].cy * tgts.height));
+        // 打印 跟踪目标 的外接矩形框的宽度、高度，w，h的值域为(0, 1]，以及w，h的像素值
+        printf("  Tracking Size (w, h) = (%.3f, %.3f), in Pixels = ((%d, %d))\n",
+          tgts.targets[0].w, tgts.targets[0].h,
+          int(tgts.targets[0].w * tgts.width),
+          int(tgts.targets[0].h * tgts.height));
+        // 打印 跟踪目标 的视线角，跟相机视场相关
+        printf("  Tracking Line-of-sight (ax, ay) = (%.3f, %.3f)\n", tgts.targets[0].los_ax, tgts.targets[0].los_ay);
+         
+          for (int i=0; i<tgts.targets.size(); i++)
+      {
+        printf("Frame-[%d], Aruco-[%d]\n", frame_id, i);
+        // 打印每个二维码的中心位置，cx，cy的值域为[0, 1]，以及cx，cy的像素值
+        printf("  Aruco Center (cx, cy) = (%.3f, %.3f), in Pixels = ((%d, %d))\n",
+          tgts.targets[i].cx, tgts.targets[i].cy,
+          int(tgts.targets[i].cx * tgts.width),
+          int(tgts.targets[i].cy * tgts.height));
+        // 打印每个二维码的外接矩形框的宽度、高度，w，h的值域为(0, 1]，以及w，h的像素值
+        printf("  Aruco Size (w, h) = (%.3f, %.3f), in Pixels = ((%d, %d))\n",
+          tgts.targets[i].w, tgts.targets[i].h,
+          int(tgts.targets[i].w * tgts.width),
+          int(tgts.targets[i].h * tgts.height));
+        // 打印每个二维码的方位角，值域为[-180, 180]
+        printf("  Aruco Yaw-angle = %.2f\n", tgts.targets[i].yaw_a);
+        // 打印每个二维码的类别，字符串类型，"aruco-?"
+        printf("  Aruco Category = %s\n", tgts.targets[i].category.c_str());
+        // 打印每个二维码的ID号
+        printf("  Aruco Tracked-ID = %d\n", tgts.targets[i].tracked_id);
+        // 打印每个二维码的视线角，跟相机视场相关
+        printf("  Aruco Line-of-sight (ax, ay) = (%.3f, %.3f)\n", tgts.targets[i].los_ax, tgts.targets[i].los_ay);
+        // 打印每个二维码的3D位置（在相机坐标系下），跟二维码实际边长、相机参数相关
+        printf("  Aruco Position = (x, y, z) = (%.3f, %.3f, %.3f)\n", tgts.targets[i].px, tgts.targets[i].py, tgts.targets[i].pz);
+      }
+      }
+
+    }
+    
+    // 显示检测结果img
+    cv::imshow(RGB_WINDOW, img);
+    cv::waitKey(1);
+  }
+
+  return 0;
+}
+
+void onMouse(int event, int x, int y, int, void*)
+{
+  if (b_clicked)
+  {
+    // 更新框选区域坐标
+    rect_sel.x = MIN(pt_origin.x, x);        
+    rect_sel.y = MIN(pt_origin.y, y);
+    rect_sel.width = abs(x - pt_origin.x);   
+    rect_sel.height = abs(y - pt_origin.y);
+  }
+  // 左键按下
+  if (event == cv::EVENT_LBUTTONDOWN)
+  {
+    b_begin_TRACK = false;  
+    b_clicked = true; 
+    pt_origin = cv::Point(x, y);       
+    rect_sel = cv::Rect(x, y, 0, 0);  
+  }
+  // 左键松开
+  else if (event == cv::EVENT_LBUTTONUP)
+  {
+    // 框选区域需要大于8x8像素
+    if (rect_sel.width * rect_sel.height < 64)
+    {
+      ;
+    }
+    else
+    {
+      b_clicked = false;
+      b_renew_ROI = true;    
+    } 
+  }
+}
diff --git a/samples/demo/camera_reading.cpp b/samples/demo/camera_reading.cpp
index 16d085d..94f3187 100644
--- a/samples/demo/camera_reading.cpp
+++ b/samples/demo/camera_reading.cpp
@@ -8,9 +8,9 @@ using namespace std;
 int main(int argc, char *argv[]) {  
   // 打开摄像头
   sv::Camera cap;
-  // cap.setWH(640, 480);
-  // cap.setFps(30);
-  cap.open(sv::CameraType::WEBCAM, 0);  // CameraID 0
+  cap.setWH(1280, 720);
+  cap.setFps(30);
+  cap.open(sv::CameraType::WEBCAM, 2);  // CameraID 0
   // 实例化OpenCV的Mat类，用于内存单帧图像
   cv::Mat img;
   while (1)
@@ -20,7 +20,7 @@ int main(int argc, char *argv[]) {
 
     // 显示img
     cv::imshow("img", img);
-    cv::waitKey(10);
+    cv::waitKey(1);
   }
 
   return 0;
diff --git a/samples/demo/car_detection_with_tracking.cpp b/samples/demo/car_detection_with_tracking.cpp
new file mode 100644
index 0000000..7b9ddc9
--- /dev/null
+++ b/samples/demo/car_detection_with_tracking.cpp
@@ -0,0 +1,193 @@
+#include <iostream>
+#include <string>
+// 包含SpireCV SDK头文件
+#include <sv_world.h>
+
+using namespace std;
+
+// 定义窗口名称
+static const std::string RGB_WINDOW = "Image window";
+// 框选到的矩形
+cv::Rect rect_sel;
+// 框选起始点
+cv::Point pt_origin;
+
+// 是否得到一个新的框选区域
+bool b_renew_ROI = false;
+// 是否开始跟踪
+bool b_begin_TRACK = false;
+// 实现框选逻辑的回调函数
+void onMouse(int event, int x, int y, int, void*);
+
+
+struct node {
+  double x,y;
+};
+node p1,p2,p3,p4;
+node p;
+double getCross(node p1, node p2, node p) {
+  return (p2.x-p1.x)*(p.y-p1.y)-(p.x-p1.x)*(p2.y-p1.y);
+}
+bool b_clicked =false;
+bool detect_tracking =true;
+
+
+int main(int argc, char *argv[]) {
+  // 定义一个新的窗口，可在上面进行框选操作
+  cv::namedWindow(RGB_WINDOW);
+  // 设置窗口操作回调函数，该函数实现整个框选逻辑
+  cv::setMouseCallback(RGB_WINDOW, onMouse, 0);
+  // 实例化 框选目标跟踪类
+  sv::SingleObjectTracker sot;
+  // 手动导入相机参数，如果使用Amov的G1等吊舱或相机，则可以忽略该步骤，将自动下载相机参数文件
+  sot.loadCameraParams(sv::get_home() + "/SpireCV/calib_webcam_640x480.yaml");
+
+
+  sv::CommonObjectDetector cod;
+  cod.loadCameraParams(sv::get_home() + "/SpireCV/calib_webcam_640x480.yaml");
+
+
+  // 打开摄像头
+  sv::Camera cap;
+  cap.setWH(640, 480);
+  cap.setFps(30);
+  cap.open(sv::CameraType::WEBCAM, 2);  // CameraID 0
+  // cv::VideoCapture cap("/home/amov/SpireCV/test/tracking_1280x720.mp4");
+  // 实例化OpenCV的Mat类，用于内存单帧图像
+  cv::Mat img;
+  int frame_id = 0;
+  while (1)
+  {
+    if (detect_tracking == true) {
+      // 实例化SpireCV的 单帧检测结果 接口类 TargetsInFrame
+      sv::TargetsInFrame tgts(frame_id++);
+      // 读取一帧图像到img
+      cap.read(img);
+      cv::resize(img, img, cv::Size(cod.image_width, cod.image_height));
+
+      // 执行通用目标检测
+      cod.detect(img, tgts);
+
+      // 可视化检测结果，叠加到img上
+      sv::drawTargetsInFrame(img, tgts);
+
+      // 控制台打印通用目标检测结果
+      printf("Frame-[%d]\n", frame_id);
+      // 打印当前检测的FPS
+      printf("  FPS = %.2f\n", tgts.fps);
+      // 打印当前相机的视场角（degree）
+      printf("  FOV (fx, fy) = (%.2f, %.2f)\n", tgts.fov_x, tgts.fov_y);
+      for (int i=0; i<tgts.targets.size(); i++)
+      {
+        printf("Frame-[%d], Object-[%d]\n", frame_id, i);
+        // 打印每个目标的中心位置，cx，cy的值域为[0, 1] 
+        printf("  Object Center (cx, cy) = (%.3f, %.3f)\n", tgts.targets[i].cx, tgts.targets[i].cy);
+        // 打印每个目标的外接矩形框的宽度、高度，w，h的值域为(0, 1] 
+        printf("  Object Size (w, h) = (%.3f, %.3f)\n", tgts.targets[i].w, tgts.targets[i].h);
+        // 打印每个目标的置信度
+        printf("  Object Score = %.3f\n", tgts.targets[i].score);
+        // 打印每个目标的类别，字符串类型
+        printf("  Object Category = %s, Category ID = [%d]\n", tgts.targets[i].category.c_str(), tgts.targets[i].category_id);
+        // 打印每个目标的视线角，跟相机视场相关
+        printf("  Object Line-of-sight (ax, ay) = (%.3f, %.3f)\n", tgts.targets[i].los_ax, tgts.targets[i].los_ay);
+        // 打印每个目标的3D位置（在相机坐标系下），跟目标实际长宽、相机参数相关
+        printf("  Object Position = (x, y, z) = (%.3f, %.3f, %.3f)\n", tgts.targets[i].px, tgts.targets[i].py, tgts.targets[i].pz);
+        p1.x = tgts.targets[i].cx * tgts.width  - tgts.targets[i].w * tgts.width  / 2;
+        p1.y = tgts.targets[i].cy * tgts.height - tgts.targets[i].h * tgts.height / 2;
+        p2.x = tgts.targets[i].cx * tgts.width  + tgts.targets[i].w * tgts.width  / 2;
+        p2.y = tgts.targets[i].cy * tgts.height - tgts.targets[i].h * tgts.height / 2;
+        p4.x = tgts.targets[i].cx * tgts.width  - tgts.targets[i].w * tgts.width  / 2;
+        p4.y = tgts.targets[i].cy * tgts.height + tgts.targets[i].h * tgts.height / 2;
+        p3.x = tgts.targets[i].cx * tgts.width  + tgts.targets[i].w * tgts.width  / 2;
+        p3.y = tgts.targets[i].cy * tgts.height + tgts.targets[i].h * tgts.height / 2;
+        p.x = pt_origin.x;
+        p.y = pt_origin.y;
+        std::cout << "p.x " << p.x << "\t" << "p.y " << p.y << std::endl;
+        if (getCross(p1, p2, p) * getCross(p3, p4, p) >= 0 && getCross(p2, p3, p) * getCross(p4, p1, p) >= 0) {
+          b_begin_TRACK = false;  
+          detect_tracking = false;
+          // pt_origin = cv::Point(nor_x, nor_p_y); 
+          // std::cout << "pt_origin  " <<nor_x<<"/t"<<nor_p_y<< std::endl;      
+          rect_sel = cv::Rect(p1.x, p1.y, tgts.targets[i].w * tgts.width, tgts.targets[i].h * tgts.height);  
+          // std::cout << rect_sel << std::endl;
+          b_renew_ROI = true;
+          frame_id = 0;
+          printf("rect_sel Yes\n");
+        }
+        else {
+          printf("rect_sel No\n");
+        }
+      }
+    }
+    else {
+      // 实例化SpireCV的 单帧检测结果 接口类 TargetsInFrame
+      sv::TargetsInFrame tgts(frame_id++);
+      // 读取一帧图像到img
+      cap.read(img);
+      cv::resize(img, img, cv::Size(sot.image_width, sot.image_height));
+      // 开始 单目标跟踪 逻辑
+      // 是否有新的目标被手动框选
+      if (b_renew_ROI)
+      {
+        // 拿新的框选区域 来 初始化跟踪器
+        sot.init(img, rect_sel);
+        // std::cout << rect_sel << std::endl;
+        // 重置框选标志
+        b_renew_ROI = false;
+        // 开始跟踪
+        b_begin_TRACK = true;
+      }
+      else if (b_begin_TRACK)
+      {
+        // 以前一帧的结果继续跟踪
+        sot.track(img, tgts);
+        // 可视化检测结果，叠加到img上
+        sv::drawTargetsInFrame(img, tgts);
+        // 控制台打印 单目标跟踪 结果
+        printf("Frame-[%d]\n", frame_id);
+        // 打印当前检测的FPS
+        printf("  FPS = %.2f\n", tgts.fps);
+        // 打印当前相机的视场角（degree）
+        printf("  FOV (fx, fy) = (%.2f, %.2f)\n", tgts.fov_x, tgts.fov_y);
+        if (tgts.targets.size() > 0)
+        {
+          printf("Frame-[%d]\n", frame_id);
+          // 打印 跟踪目标 的中心位置，cx，cy的值域为[0, 1] 
+          printf("  Tracking Center (cx, cy) = (%.3f, %.3f)\n", tgts.targets[0].cx, tgts.targets[0].cy);
+          // 打印 跟踪目标 的外接矩形框的宽度、高度，w，h的值域为(0, 1] 
+          printf("  Tracking Size (w, h) = (%.3f, %.3f)\n", tgts.targets[0].w, tgts.targets[0].h);
+          // 打印 跟踪目标 的视线角，跟相机视场相关
+          printf("  Tracking Line-of-sight (ax, ay) = (%.3f, %.3f)\n", tgts.targets[0].los_ax, tgts.targets[0].los_ay);
+        }
+      }
+    }//end of tracking     
+    // 显示检测结果img
+    cv::imshow(RGB_WINDOW, img);
+    cv::waitKey(1);
+  }
+  return 0;
+}
+
+void onMouse(int event, int x, int y, int, void*)
+{
+  if (b_clicked)
+  {
+    // 更新框选区域坐标
+    pt_origin.x = 0;
+    pt_origin.y = 0;
+  }
+  // 左键按下
+  if (event == cv::EVENT_LBUTTONDOWN)
+  {
+    detect_tracking = true;
+    pt_origin = cv::Point(x, y);  
+  }
+
+  else if (event == cv::EVENT_RBUTTONDOWN)
+  {
+    detect_tracking = true;
+    b_renew_ROI = false;
+    b_begin_TRACK = false;
+    b_clicked = true;
+  }
+}
diff --git a/samples/demo/common_object_detection.cpp b/samples/demo/common_object_detection.cpp
index f051756..544cad2 100644
--- a/samples/demo/common_object_detection.cpp
+++ b/samples/demo/common_object_detection.cpp
@@ -13,9 +13,9 @@ int main(int argc, char *argv[]) {
   
   // 打开摄像头
   sv::Camera cap;
-  // cap.setWH(640, 480);
-  // cap.setFps(30);
-  cap.open(sv::CameraType::WEBCAM, 0);  // CameraID 0
+  cap.setWH(640, 480);
+  cap.setFps(60);
+  cap.open(sv::CameraType::WEBCAM, 2);  // CameraID 0
   // 实例化OpenCV的Mat类，用于内存单帧图像
   cv::Mat img;
   int frame_id = 0;
@@ -65,7 +65,7 @@ int main(int argc, char *argv[]) {
     
     // 显示检测结果img
     cv::imshow("img", img);
-    cv::waitKey(10);
+    cv::waitKey(1);
   }
 
   return 0;
diff --git a/samples/demo/detection_with_clicked_tracking.cpp b/samples/demo/detection_with_clicked_tracking.cpp
index 7cefc72..e9c9285 100644
--- a/samples/demo/detection_with_clicked_tracking.cpp
+++ b/samples/demo/detection_with_clicked_tracking.cpp
@@ -51,7 +51,7 @@ int main(int argc, char *argv[]) {
   sv::Camera cap;
   // cap.setWH(640, 480);
   // cap.setFps(30);
-  cap.open(sv::CameraType::WEBCAM, 0);  // CameraID 0
+  cap.open(sv::CameraType::WEBCAM, 2);  // CameraID 0
   // cv::VideoCapture cap("/home/amov/SpireCV/test/tracking_1280x720.mp4");
   // 实例化OpenCV的Mat类，用于内存单帧图像
   cv::Mat img;
diff --git a/samples/demo/ellipse_detection.cpp b/samples/demo/ellipse_detection.cpp
index 596bf2f..d427c47 100644
--- a/samples/demo/ellipse_detection.cpp
+++ b/samples/demo/ellipse_detection.cpp
@@ -13,8 +13,8 @@ int main(int argc, char *argv[]) {
   
   // 打开摄像头
   sv::Camera cap;
-  // cap.setWH(640, 480);
-  // cap.setFps(30);
+  cap.setWH(640, 480);
+  cap.setFps(30);
   cap.open(sv::CameraType::WEBCAM, 0);  // CameraID 0
   // 实例化OpenCV的Mat类，用于内存单帧图像
   cv::Mat img;
@@ -63,7 +63,7 @@ int main(int argc, char *argv[]) {
     
     // 显示检测结果img
     cv::imshow("img", img);
-    cv::waitKey(10);
+    cv::waitKey(1);
   }
 
   return 0;
diff --git a/sv_algorithm_params.json b/sv_algorithm_params.json
index 00abb3a..2d4edb1 100644
--- a/sv_algorithm_params.json
+++ b/sv_algorithm_params.json
@@ -1,11 +1,23 @@
 {
     "CommonObjectDetector": {
+<<<<<<< Updated upstream
         "dataset": "COCO",
         "inputSize": 1280,
         "nmsThrs": 0.6,
         "scoreThrs": 0.3,
         "useWidthOrHeight": 1,
         "withSegmentation": false,
+=======
+        "dataset": "CAR",
+        "inputSize": 640,
+        "nmsThrs": 0.6,
+        "scoreThrs": 0.4,
+        "useWidthOrHeight": 1,
+        "withSegmentation": false,
+        "datasetCAR":{
+           "diff_car":[0.12,0.105]
+        },
+>>>>>>> Stashed changes
         "datasetPersonVehicle": {
             "person": [0.5, 1.8],
             "car": [4.1, 1.5],
@@ -114,6 +126,7 @@
     "SingleObjectTracker": {
         "algorithm": "nano",
         "backend": 0,
+<<<<<<< Updated upstream
         "target": 0,
         "useWidthOrHeight": 0,
         "sigleobjectW":2,
@@ -125,6 +138,16 @@
     },
     "EllipseDetector": {
         "radiusInMeter": 0.1535,
+=======
+        "target": 0
+    },
+    "LandingMarkerDetector": {
+        "labels": ["h"],
+        "maxCandidates": 5
+    },
+    "EllipseDetector": {
+        "radiusInMeter": 0.685,
+>>>>>>> Stashed changes
         "preProcessingGaussKernel": 5,
         "preProcessingGaussSigma": 1.306,
         "thPosition": 1.0,
@@ -132,10 +155,17 @@
         "minEdgeLength": 9,
         "minOrientedRectSide": 2.984,
         "distanceToEllipseContour": 0.111,
+<<<<<<< Updated upstream
         "minScore": 0.7,
         "minReliability": 0.470,
         "ns": 22,
         "percentNe": 0.99,
+=======
+        "minScore": 0.511,
+        "minReliability": 0.470,
+        "ns": 22,
+        "percentNe": 0.946,
+>>>>>>> Stashed changes
         "T_CNC": 0.121,
         "T_TCN_L": 0.468,
         "T_TCN_P": 0.560,
@@ -144,8 +174,13 @@
     "ArucoDetector": {
         "dictionaryId": 10,
         "markerIds": [-1],
+<<<<<<< Updated upstream
         "markerLengths": [0.2],
         "adaptiveThreshConstant": 35,
+=======
+        "markerLengths": [0.17],
+        "adaptiveThreshConstant": 7,
+>>>>>>> Stashed changes
         "adaptiveThreshWinSizeMax": 23,
         "adaptiveThreshWinSizeMin": 3,
         "adaptiveThreshWinSizeStep": 10, 
diff --git a/sv_algorithm_params_coco_1280.json b/sv_algorithm_params_coco_1280.json
new file mode 100644
index 0000000..13dc7d1
--- /dev/null
+++ b/sv_algorithm_params_coco_1280.json
@@ -0,0 +1,184 @@
+{
+    "CommonObjectDetector": {
+        "dataset": "COCO",
+        "inputSize": 1280,
+        "nmsThrs": 0.6,
+        "scoreThrs": 0.4,
+        "useWidthOrHeight": 1,
+        "withSegmentation": false,
+        "datasetPersonVehicle": {
+            "person": [0.5, 1.8],
+            "car": [4.1, 1.5],
+            "bus": [10, 3],
+            "truck": [-1, -1],
+            "bike": [-1, -1],
+            "train": [-1, -1],
+            "boat": [-1, -1],
+            "aeroplane": [-1, -1]
+        },
+        "datasetDrone": {
+            "drone": [0.4, 0.2]
+        },
+        "datasetCOCO": {
+            "person": [-1, -1],
+            "bicycle": [-1, -1],
+            "car": [-1, -1],
+            "motorcycle": [-1, -1],
+            "airplane": [-1, -1],
+            "bus": [-1, -1],
+            "train": [-1, -1],
+            "truck": [-1, -1],
+            "boat": [-1, -1],
+            "traffic light": [-1, -1],
+            "fire hydrant": [-1, -1],
+            "stop sign": [-1, -1],
+            "parking meter": [-1, -1],
+            "bench": [-1, -1],
+            "bird": [-1, -1],
+            "cat": [-1, -1],
+            "dog": [-1, -1],
+            "horse": [-1, -1],
+            "sheep": [-1, -1],
+            "cow": [-1, -1],
+            "elephant": [-1, -1],
+            "bear": [-1, -1],
+            "zebra": [-1, -1],
+            "giraffe": [-1, -1],
+            "backpack": [-1, -1],
+            "umbrella": [-1, -1],
+            "handbag": [-1, -1],
+            "tie": [-1, -1],
+            "suitcase": [-1, -1],
+            "frisbee": [-1, -1],
+            "skis": [-1, -1],
+            "snowboard": [-1, -1],
+            "sports ball": [-1, -1],
+            "kite": [-1, -1],
+            "baseball bat": [-1, -1],
+            "baseball glove": [-1, -1],
+            "skateboard": [-1, -1],
+            "surfboard": [-1, -1],
+            "tennis racket": [-1, -1],
+            "bottle": [-1, -1],
+            "wine glass": [-1, -1],
+            "cup": [-1, -1],
+            "fork": [-1, -1],
+            "knife": [-1, -1],
+            "spoon": [-1, -1],
+            "bowl": [-1, -1],
+            "banana": [-1, -1],
+            "apple": [-1, -1],
+            "sandwich": [-1, -1],
+            "orange": [-1, -1],
+            "broccoli": [-1, -1],
+            "carrot": [-1, -1],
+            "hot dog": [-1, -1],
+            "pizza": [-1, -1],
+            "donut": [-1, -1],
+            "cake": [-1, -1],
+            "chair": [-1, -1],
+            "couch": [-1, -1],
+            "potted plant": [-1, -1],
+            "bed": [-1, -1],
+            "dining table": [-1, -1],
+            "toilet": [-1, -1],
+            "tv": [-1, -1],
+            "laptop": [-1, -1],
+            "mouse": [-1, -1],
+            "remote": [-1, -1],
+            "keyboard": [-1, -1],
+            "cell phone": [-1, -1],
+            "microwave": [-1, -1],
+            "oven": [-1, -1],
+            "toaster": [-1, -1],
+            "sink": [-1, -1],
+            "refrigerator": [-1, -1],
+            "book": [-1, -1],
+            "clock": [-1, -1],
+            "vase": [-1, -1],
+            "scissors": [-1, -1],
+            "teddy bear": [-1, -1],
+            "hair drier": [-1, -1],
+            "toothbrush": [-1, -1]
+        }
+    },
+    "AutoFocusObjectDetector": {
+        "lock_thres": 5,
+        "unlock_thres": 5,
+        "lock_scale_init": 12.0,
+        "lock_scale": 8.0,
+        "categories_filter": [],
+        "keep_unlocked": false,
+        "use_square_region": false
+    },
+    "SingleObjectTracker": {
+        "algorithm": "siamrpn",
+        "backend": 0,
+        "target": 0
+    },
+    "LandingMarkerDetector": {
+        "labels": ["x", "h"],
+        "maxCandidates": 5
+    },
+    "EllipseDetector": {
+        "radiusInMeter": 0.5,
+        "preProcessingGaussKernel": 5,
+        "preProcessingGaussSigma": 1.306,
+        "thPosition": 1.0,
+        "maxCenterDistance": 0.05,
+        "minEdgeLength": 9,
+        "minOrientedRectSide": 2.984,
+        "distanceToEllipseContour": 0.111,
+        "minScore": 0.511,
+        "minReliability": 0.470,
+        "ns": 22,
+        "percentNe": 0.946,
+        "T_CNC": 0.121,
+        "T_TCN_L": 0.468,
+        "T_TCN_P": 0.560,
+        "thRadius": 0.202
+    },
+    "ArucoDetector": {
+        "dictionaryId": 10,
+        "markerIds": [-1],
+        "markerLengths": [0.2],
+        "adaptiveThreshConstant": 7,
+        "adaptiveThreshWinSizeMax": 23,
+        "adaptiveThreshWinSizeMin": 3,
+        "adaptiveThreshWinSizeStep": 10, 
+        "aprilTagCriticalRad": 0.17453292519,
+        "aprilTagDeglitch": 0,
+        "aprilTagMaxLineFitMse": 10.0,
+        "aprilTagMaxNmaxima": 10,
+        "aprilTagMinClusterPixels": 5,
+        "aprilTagMinWhiteBlackDiff": 5,
+        "aprilTagQuadDecimate": 0.0,
+        "aprilTagQuadSigma": 0.0,
+        "cornerRefinementMaxIterations": 30,
+        "cornerRefinementMethod": 0,
+        "cornerRefinementMinAccuracy": 0.1,
+        "cornerRefinementWinSize": 5,
+        "detectInvertedMarker": false,
+        "errorCorrectionRate": 0.6,
+        "markerBorderBits": 1,
+        "maxErroneousBitsInBorderRate": 0.35,
+        "maxMarkerPerimeterRate": 4.0,
+        "minCornerDistanceRate": 0.05,
+        "minDistanceToBorder": 3,
+        "minMarkerDistanceRate": 0.05,
+        "minMarkerLengthRatioOriginalImg": 0,
+        "minMarkerPerimeterRate": 0.03,
+        "minOtsuStdDev": 5.0,
+        "minSideLengthCanonicalImg": 32,
+        "perspectiveRemoveIgnoredMarginPerCell": 0.13,
+        "perspectiveRemovePixelPerCell": 4,
+        "polygonalApproxAccuracyRate": 0.03,
+        "useAruco3Detection": false
+    },
+    "ColorLineDetector": {
+        "line_color": "black",
+        "line_location": 0.5,
+        "line_location_a1": 0.3,
+        "line_location_a2": 0.7
+    }
+}
diff --git a/sv_algorithm_params_coco_640.json b/sv_algorithm_params_coco_640.json
new file mode 100644
index 0000000..8ccf8a7
--- /dev/null
+++ b/sv_algorithm_params_coco_640.json
@@ -0,0 +1,184 @@
+{
+    "CommonObjectDetector": {
+        "dataset": "COCO",
+        "inputSize": 640,
+        "nmsThrs": 0.6,
+        "scoreThrs": 0.4,
+        "useWidthOrHeight": 1,
+        "withSegmentation": true,
+        "datasetPersonVehicle": {
+            "person": [0.5, 1.8],
+            "car": [4.1, 1.5],
+            "bus": [10, 3],
+            "truck": [-1, -1],
+            "bike": [-1, -1],
+            "train": [-1, -1],
+            "boat": [-1, -1],
+            "aeroplane": [-1, -1]
+        },
+        "datasetDrone": {
+            "drone": [0.4, 0.2]
+        },
+        "datasetCOCO": {
+            "person": [-1, -1],
+            "bicycle": [-1, -1],
+            "car": [-1, -1],
+            "motorcycle": [-1, -1],
+            "airplane": [-1, -1],
+            "bus": [-1, -1],
+            "train": [-1, -1],
+            "truck": [-1, -1],
+            "boat": [-1, -1],
+            "traffic light": [-1, -1],
+            "fire hydrant": [-1, -1],
+            "stop sign": [-1, -1],
+            "parking meter": [-1, -1],
+            "bench": [-1, -1],
+            "bird": [-1, -1],
+            "cat": [-1, -1],
+            "dog": [-1, -1],
+            "horse": [-1, -1],
+            "sheep": [-1, -1],
+            "cow": [-1, -1],
+            "elephant": [-1, -1],
+            "bear": [-1, -1],
+            "zebra": [-1, -1],
+            "giraffe": [-1, -1],
+            "backpack": [-1, -1],
+            "umbrella": [-1, -1],
+            "handbag": [-1, -1],
+            "tie": [-1, -1],
+            "suitcase": [-1, -1],
+            "frisbee": [-1, -1],
+            "skis": [-1, -1],
+            "snowboard": [-1, -1],
+            "sports ball": [-1, -1],
+            "kite": [-1, -1],
+            "baseball bat": [-1, -1],
+            "baseball glove": [-1, -1],
+            "skateboard": [-1, -1],
+            "surfboard": [-1, -1],
+            "tennis racket": [-1, -1],
+            "bottle": [-1, -1],
+            "wine glass": [-1, -1],
+            "cup": [-1, -1],
+            "fork": [-1, -1],
+            "knife": [-1, -1],
+            "spoon": [-1, -1],
+            "bowl": [-1, -1],
+            "banana": [-1, -1],
+            "apple": [-1, -1],
+            "sandwich": [-1, -1],
+            "orange": [-1, -1],
+            "broccoli": [-1, -1],
+            "carrot": [-1, -1],
+            "hot dog": [-1, -1],
+            "pizza": [-1, -1],
+            "donut": [-1, -1],
+            "cake": [-1, -1],
+            "chair": [-1, -1],
+            "couch": [-1, -1],
+            "potted plant": [-1, -1],
+            "bed": [-1, -1],
+            "dining table": [-1, -1],
+            "toilet": [-1, -1],
+            "tv": [-1, -1],
+            "laptop": [-1, -1],
+            "mouse": [-1, -1],
+            "remote": [-1, -1],
+            "keyboard": [-1, -1],
+            "cell phone": [-1, -1],
+            "microwave": [-1, -1],
+            "oven": [-1, -1],
+            "toaster": [-1, -1],
+            "sink": [-1, -1],
+            "refrigerator": [-1, -1],
+            "book": [-1, -1],
+            "clock": [-1, -1],
+            "vase": [-1, -1],
+            "scissors": [-1, -1],
+            "teddy bear": [-1, -1],
+            "hair drier": [-1, -1],
+            "toothbrush": [-1, -1]
+        }
+    },
+    "AutoFocusObjectDetector": {
+        "lock_thres": 5,
+        "unlock_thres": 5,
+        "lock_scale_init": 12.0,
+        "lock_scale": 8.0,
+        "categories_filter": [],
+        "keep_unlocked": false,
+        "use_square_region": false
+    },
+    "SingleObjectTracker": {
+        "algorithm": "siamrpn",
+        "backend": 0,
+        "target": 0
+    },
+    "LandingMarkerDetector": {
+        "labels": ["x", "h"],
+        "maxCandidates": 5
+    },
+    "EllipseDetector": {
+        "radiusInMeter": 0.5,
+        "preProcessingGaussKernel": 5,
+        "preProcessingGaussSigma": 1.306,
+        "thPosition": 1.0,
+        "maxCenterDistance": 0.05,
+        "minEdgeLength": 9,
+        "minOrientedRectSide": 2.984,
+        "distanceToEllipseContour": 0.111,
+        "minScore": 0.511,
+        "minReliability": 0.470,
+        "ns": 22,
+        "percentNe": 0.946,
+        "T_CNC": 0.121,
+        "T_TCN_L": 0.468,
+        "T_TCN_P": 0.560,
+        "thRadius": 0.202
+    },
+    "ArucoDetector": {
+        "dictionaryId": 10,
+        "markerIds": [-1],
+        "markerLengths": [0.2],
+        "adaptiveThreshConstant": 7,
+        "adaptiveThreshWinSizeMax": 23,
+        "adaptiveThreshWinSizeMin": 3,
+        "adaptiveThreshWinSizeStep": 10, 
+        "aprilTagCriticalRad": 0.17453292519,
+        "aprilTagDeglitch": 0,
+        "aprilTagMaxLineFitMse": 10.0,
+        "aprilTagMaxNmaxima": 10,
+        "aprilTagMinClusterPixels": 5,
+        "aprilTagMinWhiteBlackDiff": 5,
+        "aprilTagQuadDecimate": 0.0,
+        "aprilTagQuadSigma": 0.0,
+        "cornerRefinementMaxIterations": 30,
+        "cornerRefinementMethod": 0,
+        "cornerRefinementMinAccuracy": 0.1,
+        "cornerRefinementWinSize": 5,
+        "detectInvertedMarker": false,
+        "errorCorrectionRate": 0.6,
+        "markerBorderBits": 1,
+        "maxErroneousBitsInBorderRate": 0.35,
+        "maxMarkerPerimeterRate": 4.0,
+        "minCornerDistanceRate": 0.05,
+        "minDistanceToBorder": 3,
+        "minMarkerDistanceRate": 0.05,
+        "minMarkerLengthRatioOriginalImg": 0,
+        "minMarkerPerimeterRate": 0.03,
+        "minOtsuStdDev": 5.0,
+        "minSideLengthCanonicalImg": 32,
+        "perspectiveRemoveIgnoredMarginPerCell": 0.13,
+        "perspectiveRemovePixelPerCell": 4,
+        "polygonalApproxAccuracyRate": 0.03,
+        "useAruco3Detection": false
+    },
+    "ColorLineDetector": {
+        "line_color": "black",
+        "line_location": 0.5,
+        "line_location_a1": 0.3,
+        "line_location_a2": 0.7
+    }
+}