Migrate examples to use depthai 2.0.0.1 and follow a common blob naming format #96
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
While migrating examples to the latest PyPi release, and making sure the blobs are using the latest OpenVINO and have correct blob names, I found two examples that stopped working as they used to:
Pedestrian Reidentification network (
person-reidentification-retail-0031
) is no longer available in 2021.2, was removed in 2020.4. The new network that I tried to use - person_reidentification_retail_0288 and person_reidentification_retail_0277 - accept an input image in higher resolution (now128x256
, previously48x96
), which may be a cause of a significant inference speed drop when these networks start to process the frames.Code for this example is here as was not included in this PR due to the performance issues
Tensorflow Image Classification example has accuracy issues when model is changed to the one compiled for 6 or 8 shaves. Where previously is recognized a dandelion (like on a demo image) it does no more. That's why this example is also not included in this PR, and its source code is here
Other than these two, the rest of the examples work fine.