Why there is no bounding box for YOLOV3 COCO Detection Video?

Hi,you should refer to convert model.

https://gitee.com/ascend/samples/blob/master/cplusplus/level2_simple_inference/2_object_detection/YOLOV3_coco_detection_video/README.md

1. Run the following commands to download the AIPP configuration file and convert the model:

   cd $HOME/models/YOLOV3_coco_detection_video

   wget https://c7xcode.obs.cn-north-4.myhuaweicloud.com/models/YOLOV3_coco_detection_video/aipp_bgr.cfg

   atc --model=./yolov3.prototxt --weight=./yolov3.caffemodel --framework=0 --output=yolov3 --soc_version=Ascend310 --insert_op_conf=./aipp_bgr.cfg

2. Run the following command to copy the converted model to the model folder of the sample:

   cp ./yolov3.om $HOME/samples/cplusplus/level2_simple_inference/2_object_detection/YOLOV3_coco_detection_video/model/

Dear XYT,

I did download model and ATC things, following instruction from https://gitee.com/ascend/modelzoo/tree/master/contrib/TensorFlow/Research/cv/yolov3/ATC_yolov3_caffe_AE

that mention

atc --model=yolov3.prototxt --weight=yolov3.caffemodel --framework=0 --output=yolov3_framework_caffe_aipp_1_batch_1_input_int8_output_FP32.om --soc_version=Ascend310 --insert_op_conf=aipp_nv12.cfg

I noticed that you mention ATC using aipp_bgr.cfg

wget 

https://c7xcode.obs.cn-north-4.myhuaweicloud.com/models/YOLOV3_coco_detection_video/aipp_bgr.cfg

It works!

So, please tell me what is the different between: aipp_nv12.cfg and aipp_bgr.cfg?

How do we configure it correctly?