Add OpenVINO model problem in VOE (factory-ai-vision) : When I replace other model and use the same name , This model will not work in Deployment(no any bounding box)

[Henry1Chen]

Please provide us with the following information:

This issue is for a: (mark with an x)

- [ x ] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

Minimal steps to reproduce

Please follow below Linux command.

1. Go into module menager docker.
   --docker exec -u 0 -it modelmanager bash
2. Go to downloader
   --cd /app/downloader/tools/downloader
3. Setup environment
   --./setup.py install
4. Download model
   --./downloader.py --name face-detection-retail-0005
5. Copy model to /workspace loaction.
   --cp /app/downloader/tools/downloader/intel/face-detection-retail-0005/FP32/face-detection-retail-0005.bin /workspace/face- detection-retail-0004/1/
   --cp /app/downloader/tools/downloader/intel/face-detection-retail-0005/FP32/face-detection-retail-0005.xml /workspace/face-detection-retail-0004/1/
6. Remove face-detection-retail-0004 .bin and .xml file
   --rm /workspace/face-detection-retail-0004/1/face-detection-retail-0004.xml
   --rm /workspace/face-detection-retail-0004/1/face-detection-retail-0004.bin
7. Change face-detection-retail-0005 .bin and .xml name face-detection-retail-0004 .bin and .xml
   --cd /workspace/face-detection-retail-0004/1
   --mv face-detection-retail-0005.bin face-detection-retail-0004.bin
   --mv face-detection-retail-0005.xml face-detection-retail-0004.xml

Any log messages given by the failure

log will show “Validation failed Node: Face Detection has no input with name: data” below in ovmsserver docker container.

Expected/desired behavior

Deployment will show objection detection successful and not show any error log.

OS and Version?

Linux ubuntu 20.04

Versions

Commit version :
OpenVINO model problem in VOE.pptx
e47397d

Mention any other details that might be useful

The source is in factory-ai-vision

---

Thanks! We'll be in touch soon.