How to draw bounding boxes around classified objects using tensorflow lite?

Question

I would like to know if it is possible to draw bounding boxes using Tensorflow lite. I have been able to draw them using tensorflow-android in version 1.12 but I have no example for drawing bounding boxes in tensorflow lite.

In the code below you can see my way in tensorflow-android 1.12 to get the outputLocations which is working.

 inferenceInterface.run(outputNames, logStats);
LOGGER.d("End Section run " + System.currentTimeMillis());
Trace.endSection();

// Copy the output Tensor back into the output array.
Trace.beginSection("fetch");
LOGGER.d("Begin Section fetch " + System.currentTimeMillis());
outputLocations = new float[MAX_RESULTS * 4];
outputScores = new float[MAX_RESULTS];
outputClasses = new float[MAX_RESULTS];
outputNumDetections = new float[1];
inferenceInterface.fetch(outputNames[0], outputLocations);

It would be great if you could tell me how to get outputLocations using runInference() from trensorflow-lite instead.

Answer 1

如果您使用以下对象检测模型： http : //download.tensorflow.org/models/object_detection/ssd_mobilenet_v1_quantized_300x300_coco14_sync_2018_07_18.tar.gz输出张量已经具有输出位置，分数，类等。您可以遵循类似于Android的示例Java示例应用程序： https : //github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/examples/android/app/src/main/java/org/tensorflow/demo/TFLiteObjectDetectionAPIModel.java