[ViPPET] Add new retail predefined pipelines #1607
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… counting and replacement
…tection with new input files and descriptions
This reverts commit 151202e.
…unter to the end to measure end-to-end FPS
…ndled by DLStreamer's gvafpscounter
…iction in definitions
ktaube26
requested review from
dlemiech,
dmichalo,
grey-eminence,
msmiatac and
pmalatyn
as code owners
Contributor
There was a problem hiding this comment.
Pull request overview
This pull request refactors the pipeline loading mechanism to use individual .yaml config files instead of directories and adds several new retail analytics pipeline configurations for CPU, GPU, and NPU. It also updates tests, removes unused scripts, and adds video resources.
Changes:
- Refactored
PipelineLoaderto work with.yamlfiles directly instead of pipeline directories, simplifying discovery and configuration loading - Added 9 new retail pipeline configuration files for face detection/age-gender recognition, YOLO 11n object detection, and EfficientNet B0 classification across CPU, GPU, and NPU
- Updated test suite to reflect file-based pipeline loading
- Removed 3 unused retail-related Python scripts and added 2 new video resources for testing
Reviewed changes
Copilot reviewed 20 out of 36 changed files in this pull request and generated 2 comments.
Show a summary per file
| File | Description |
|---|---|
| vippet/pipelines/loader.py | Refactored to list and load pipeline configs from .yaml files instead of directories |
| vippet/managers/pipeline_manager.py | Updated to use config paths instead of pipeline names |
| vippet/tests/loader_test.py | New test file for file-based pipeline loader |
| vippet/tests/loader.py | Deleted old directory-based loader tests |
| vippet/video_encoder.py | Enhanced fakesink replacement logic to handle embedded property values |
| vippet/pipelines/*.yaml | Added 9 new retail pipeline configurations for various device types |
| ui/src/pages/Home.tsx | Added image imports for new pipelines |
| shared/videos/default_recordings.yaml | Added video resources for object classification and age prediction |
| shared/scripts/*.py | Removed unused barcode and OCR processing scripts |
| shared/models/supported_models.yaml | Enabled NPU support for efficientnet-b0 and yolo11n models |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
...tform-evaluation-tool/vippet/pipelines/retail-face-detection-age-gender-recognition-gpu.yaml
Outdated
Show resolved
Hide resolved
...sual-pipeline-and-platform-evaluation-tool/vippet/pipelines/retail-yolo11n-effnetb0-gpu.yaml
Outdated
Show resolved
Hide resolved
tools/visual-pipeline-and-platform-evaluation-tool/vippet/tests/loader_test.py
Outdated
Show resolved
Hide resolved
KaliszWiktoria
previously approved these changes
Jan 13, 2026
p-zak
requested changes
Jan 13, 2026
tools/visual-pipeline-and-platform-evaluation-tool/vippet/pipelines/loader.py
Outdated
Show resolved
Hide resolved
tools/visual-pipeline-and-platform-evaluation-tool/vippet/pipelines/loader.py
Outdated
Show resolved
Hide resolved
p-zak
approved these changes
Jan 14, 2026
KaliszWiktoria
approved these changes
Jan 14, 2026
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Description
This pull request refactors the pipeline loading mechanism to use individual
.yamlconfig files for pipelines instead of directories, and adds several new retail analytics pipeline configurations for both CPU and GPU. It also updates the test suite to reflect these changes and adds a new video resource for object classification. The most important changes are grouped below:Pipeline Loader Refactor:
PipelineLoaderinloader.pyto list pipeline config.yamlfiles instead of directories, removed the pipeline name validation logic, and updated the config loading method to accept file paths directly. This simplifies pipeline discovery and loading.PipelineManagerto work with the new config path-based pipeline loading.New Retail Pipeline Configurations:
Testing Updates:
tests/loader.pywith a new one intests/loader_test.pyto test the new file-based pipeline loader logic. [1] [2]Video Resource Update:
default_recordings.yamlto support the new retail pipelines.Checklist: