This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TDA2EXEVM: question about the jDetNetV2 result after importing from deploy.prototxt and .caffemodel

Part Number: TDA2EXEVM

Hello!

By following the instruction of how to use tidl for object detection, I download the deploy.prototxt and .caffemodel from https://github.com/tidsp/caffe-jacinto-models/tree/caffe-0.16/trained/object_detection/voc0712/JDetNet/ssd768x320_ds_PSP_dsFac_32_hdDS8_0/sparse

Then I use the import Tool tidl_model_import.out.exe provided from\ti_components\algorithms\REL.TIDL.01.01.03.00\modules\ti_dl\utils\importTool to get net.bin and param.bin. 

Next I create a usecase to test the result offline (use NullSource to feed in the test video). At the beginning of the video (about 30s), there is no object detected. But after 30s, the thing goes right. Could you explain the reason for this phenomenon? Or in which step I make a mistake maybe? And how could I fix it? Thank you!