Tensorflow 对象检测 API 索引越界 [英] Tensorflow Object Detection API Index out of Bounds

问题描述

在运行 Android TF 检测演示并使用我重新训练的 3 类 ssd_mobilenet_v1_coco 模型后，我的 TF 检测演示崩溃并给出 IndexOutOfBoundException，即

Upon running the Android TF Detect Demo and using my retrained 3 class ssd_mobilenet_v1_coco model my TF detect Demo crashes giving an IndexOutOfBoundException which is

12-26 17:53:13.931 22429-25212/org.tensorflow.demo E/AndroidRuntime:致命异常:推理进程:org.tensorflow.demo，PID:22429java.lang.ArrayIndexOutOfBoundsException: length=3;指数=3在 java.util.Vector.arrayIndexOutOfBoundsException(Vector.java:907)在 java.util.Vector.elementAt(Vector.java:328)在 java.util.Vector.get(Vector.java:442)在 org.tensorflow.demo.TensorFlowObjectDetectionAPIModel.recognizeImage(TensorFlowObjectDetectionAPIModel.java:194)在 org.tensorflow.demo.DetectorActivity$3.run(DetectorActivity.java:289)在 android.os.Handler.handleCallback(Handler.java:739)在 android.os.Handler.dispatchMessage(Handler.java:95)在 android.os.Looper.loop(Looper.java:148)在 android.os.HandlerThread.run(HandlerThread.java:61)

12-26 17:53:13.931 22429-25212/org.tensorflow.demo E/AndroidRuntime: FATAL EXCEPTION: inference Process: org.tensorflow.demo, PID: 22429 java.lang.ArrayIndexOutOfBoundsException: length=3; index=3 at java.util.Vector.arrayIndexOutOfBoundsException(Vector.java:907) at java.util.Vector.elementAt(Vector.java:328) at java.util.Vector.get(Vector.java:442) at org.tensorflow.demo.TensorFlowObjectDetectionAPIModel.recognizeImage(TensorFlowObjectDetectionAPIModel.java:194) at org.tensorflow.demo.DetectorActivity$3.run(DetectorActivity.java:289) at android.os.Handler.handleCallback(Handler.java:739) at android.os.Handler.dispatchMessage(Handler.java:95) at android.os.Looper.loop(Looper.java:148) at android.os.HandlerThread.run(HandlerThread.java:61)

来自

for (int i = 0; i <= outputScores.length; ++i) {
  final RectF detection =
      new RectF(
          outputLocations[4 * i + 1] * inputSize,
          outputLocations[4 * i] * inputSize,
          outputLocations[4 * i + 3] * inputSize,
          outputLocations[4 * i + 2] * inputSize);

  pq.add(new Recognition("" + i, labels.get((int) outputClasses[i]), outputScores[i], detection));
}

推荐答案

通过添加???"解决我的 Label 文件中的字符串，看起来 App 遵循 pbtxt 格式，第一个 ID 为 1.

Solved by Adding a "???" string inside my Label file, it seems that the App follows the pbtxt format of having the first ID which is 1.

这篇关于Tensorflow 对象检测 API 索引越界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！