Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.

tensorfrow inception-v1 examples

idata
Employee
837 Views

Hello,

 

I am looking through the inception-v1examples and I observe that the "check" command reading the cat.jpg picture and expecting it to be of category 829.

 

mvNCCheck -s 12 output/inception-v1.meta -in=input -on=InceptionV1/Logits/Predictions/Reshape_1 -i ../../data/images/cat.jpg -id 829 -S 2 -M 128 -cs 0,1,2

 

Two things are quite puzzling to me.

 

1) if I pass a different image, e.g. 512_Phone.jpg, it is also categorized as 829, and passes the check.

 

mvNCCheck -s 12 output/inception-v1.meta -in=input -on=InceptionV1/Logits/Predictions/Reshape_1 -i ../../data/images/512_Phone.jpg -id 829 -S 2 -M 128 -cs 0,1,2

 

2) if I look at the categories file, I see on the line 829 stopwatch, stop watch, surrounded by other categories which are all not cats, and not phones.

 

So, obviously many things are not working as expected.

 

As a first step, I would appreciate if somebody points me to the correct interpretation of the category 829 in inception v1 network.

 

Further hints for how to make it work properly would also be appreciated.

 

Thanks in advance!

0 Kudos
0 Replies
Reply