2

I am currently working with the YoloV3-tiny. Repository: https://github.com/AlexeyAB/darknet

To import the network into C++ project I use OpenVINO-Toolkit. In more detail I use the following procedure to convert the network:
Converting YOLO* Models to the Intermediate Representation (IR)

This procedure carries out a conversion and an optimization to proceed with the inference.

Now, I would like to try the YoloV4 because it seems to be more effective for the purpose of the project. The problem is that OpenVINO Toolkit does not yet support this version and does not report the .json (file needed for optimization) file relative to version 4 but only up to version 3.

What has changed in terms of structure between version 3 and version 4 of the Yolo?
Can I hopefully hope that the conversion of the YoloV3-tiny (or YoloV3) is the same as the YoloV4?
Is the YoloV4 much slower than the YoloV3-tiny using only the CPU for inference?
When will the YoloV4-tiny be available?
Does anyone have information about it?

greybeard
  • 2,249
  • 8
  • 30
  • 66
Giosuè Nardi
  • 41
  • 1
  • 4

2 Answers2

6
  • The difference between YoloV4 and YoloV3 is the backbone. YoloV4 has CSPDarknet53, whilst YoloV3 has Darknet53 backbone. See https://arxiv.org/pdf/2004.10934.pdf.
  • Also, YoloV4 is not supported officially by OpenVINO. However, you can still test and validate YoloV4 on your end with some workaround. There is one way for now to run YoloV4 through OpenCV which will build network using nGraph API and then pass to Inference Engine. See https://github.com/opencv/opencv/pull/17185.
  • The key problem is the Mish activation function - there is no optimized implementation yet, which is why we have to implement it by definition with tanh and exponential functions. Unfortunately, one-to-one topology comparison shows significant performance degradation. The performance results are also available in the github link above.
Rommel_Intel
  • 1,369
  • 1
  • 4
  • 8
0

https://github.com/TNTWEN/OpenVINO-YOLOV4 This is my project based on v3's converter (darknet -> tensorflow ->IR)and i have finished the adaptation of OpenVINO Yolov4,v4-relu,v4-tiny. You could have a try. And you can use V4's IRmodel and run on v3's c++ demo directly

TNTWEN
  • 1
  • 1
  • 3
    From Review: Hi, while links are a great way of sharing knowledge, they won't really answer the question if they get broken in the future. Add to your answer the essential content of the link which answers the question. In case the content is too complex or too big to fit here, describe the general idea of the proposed solution. Remember to always keep a link reference to the original solution's website. See: [How do I write a good answer?](https://stackoverflow.com/help/how-to-answer) – sɐunıɔןɐqɐp Aug 05 '20 at 08:03