OpenCV Android的DNN输入大小有隐含限制吗？

Question

我训练了一个Caffe模型，该模型没有指定输入大小。然后，我使用OpenCV在Android上运行了它。代码如下。

import java.util.Random;

import javafx.animation.AnimationTimer;
import javafx.application.Application;
import javafx.application.Platform;
import javafx.scene.Group;
import javafx.scene.Scene;
import javafx.scene.canvas.Canvas;
import javafx.scene.canvas.GraphicsContext;
import javafx.stage.Stage;

public class Main extends Application{
    private Point[] triangle;
    private static int points = 1000000;
    private static Point midPoint;

    private int width = 600, height = 600;

    @Override
    public void start(Stage primaryStage) {

        Canvas canvas = new Canvas(width,height);
        GraphicsContext gc = canvas.getGraphicsContext2D();

        Scene scene = new Scene(new Group(canvas));
        primaryStage.setScene(scene);
        primaryStage.show();

        Random ran = new Random();
        triangle = new Point[3];
        triangle[1] = new Point(0d, 0d);
        triangle[0] = new Point(width/2d, (double)height);
        triangle[2] = new Point((double)width, 0d);
        midPoint = Point.findMidTo(triangle[0], triangle[1], triangle[2]);


        AnimationTimer timer = new AnimationTimer(){
            @Override
            public void handle(long now){
                if(points-- > 0) {
                    int r = ran.nextInt(3);
                    midPoint = midPoint.findMidTo(triangle[r]);
                    gc.fillOval(midPoint.getX(), midPoint.getY(),1,1);
                }
            }
        };
        timer.start();

    }

    public static void main(String[] args) {
        launch(args);
    }    
}

正如评论中所写，输入大小似乎存在某些内在或隐式限制。 private void runEntireImageTest(Mat inY) { float scaleFactor = 1.0f / 255.0f; Scalar mean = new Scalar(0); Mat resized = new Mat(); // (550, 441) runs OK. (smaller sized runs also fine) // (551, 441), (550, 442) crash. // (441, 550) crashes Imgproc.resize(inY, resized, new Size(550, 441)); Mat segBlob = Dnn.blobFromImage(resized, scaleFactor, resized.size(), mean, false, false); mNet.setInput(segBlob); // if input size is above some value, crash will happen here Mat lastLayer = mNet.forward(); Mat outY = lastLayer.reshape(1, 1); } 运行正常，但是(550, 441)会导致(551, 441)：

SIGSEGV

我认为这不是内存问题，因为A/libc: Fatal signal 11 (SIGSEGV), code 1, fault addr 0x6d494bd744 in tid 19693 (myappname), pid 19661 (myappname)运行正常，但是(550, 441)会崩溃。是什么原因造成的？