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.
6392 Discussions

Error: Infinite bounding boxes during script run using yolo v3 + lstm + app layers custom model

Karmeo
Novice
1,587 Views

Hey! I am trying to run my own yolov3 model with lstm and spp layers in the object_detection_demo_yolov3_async.py script.
My model has been successfully frozen, and also successfully standardized for OpenVINO, but when the script is run, the following happens: the model produces an infinite number of bounding boxes with an accuracy value of 1.00. The original model works well. Please, tell me how to fix this error?

the result of the openVINO / object_detection_demo_yolov3_async.py script.the result of the openVINO / object_detection_demo_yolov3_async.py script.the result of the alexeyab / darknet.py scriptthe result of the alexeyab / darknet.py script

 

0 Kudos
12 Replies
Iffa_Intel
Moderator
1,567 Views

Hi,


If possible can you share the model that you are using & having issues with?




Sincerely,

Iffa


0 Kudos
Karmeo
Novice
1,559 Views

Hi, posting a link to working model, frozen model and IR model

link: https://disk.yandex.ru/d/aJ31p5Fn4ig8qg

0 Kudos
Iffa_Intel
Moderator
1,543 Views

Hi,

 

Your model managed to be run with the object_detection_demo.py Python demo, however there are no bounding boxes or etc.

I believe you'll need to train this model first. You may refer to my attachment.

The trained model should give out the output as my 2nd attachment. (I used yolo-v3-tf pretrained model with the demo app).

 

Another thing to take note, the object detection sample app in the latest OpenVINO is named object_detection_demo.py and you will need to use the -at (ssd,yolo,faceboxes,centernet,retinaface):

 https://docs.openvinotoolkit.org/latest/omz_demos_object_detection_demo_python.html

 

I suggest that you use the latest OpenVINO version with the latest demo/samples application.

 

Sincerely,

Iffa

 

 

 

0 Kudos
Karmeo
Novice
1,535 Views

Hello again. I am having a problem when starting a custom model with an lstm layer added. I have trained the network for custom classes (stickers). I successfully launched the classic network configuration on openvino, the problem is just when adding custom layers

I am attaching an example from the test sample

0 Kudos
Iffa_Intel
Moderator
1,522 Views

This thread might helps you:

https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/LSTM-Support-in-Openvino-Tensorflow-and-others/td-p/1205698


(LSTM is supported with some limitations)



Sincerely,

Iffa


0 Kudos
Karmeo
Novice
1,518 Views

Hello, can you tell me more, is the bottleneck and output in lstm layers attribute supported?

Showing used attributes in my model:

[conv_lstm]
bottleneck = 1
batch_normalize=1
size=3
pad=1
output=256
peephole=0

 

0 Kudos
Iffa_Intel
Moderator
1,507 Views

May I know your reference source for LSTM attributes that you had added into your model (bottleneck & output) as from this TF link, none of these attributes are added into the LSTM layer - https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTM#attributes

 

Plus, did the custom model (yolov3) is implemented in TF2 Keras?


Sincerely,

Iffa

 


0 Kudos
Karmeo
Novice
1,501 Views

Hi, I took the configuration from this repository: https://github.com/AlexeyAB/darknet/issues/3114  and taught it on my own material. Added bottleneck feature so the network can learn on an average 1060 graphics card

Снимок экрана от 2021-06-29 11-10-28.png

0 Kudos
Iffa_Intel
Moderator
1,468 Views

Thank you for your patience. This reply took quite some time due to the delay from the engineering team's perspective.


Currently, it is confirmed that the "bottleneck" attribute is not supported.


Sincerely,

Iffa



0 Kudos
Karmeo
Novice
1,456 Views

OK, thanks for answer. Can you please tell me if this function will be supported in the next versions? And do you know how you can work around this error at this time?

0 Kudos
Iffa_Intel
Moderator
1,448 Views

We will share any updates if they are available soon.



Sincerely,

Iffa


0 Kudos
Iffa_Intel
Moderator
1,427 Views

Hi,


Fyi, you can refer to this release note for updated features once the new release is out:

https://software.intel.com/content/www/us/en/develop/articles/openvino-relnotes.html


Intel will no longer monitor this thread since we have provided a solution. If you need any additional information from Intel, please submit a new question.


Sincerely,

Iffa



0 Kudos
Reply