Skip to content

feat: add video analysis operators and qwen vl service deployment files#469

Open
guotingxuan wants to merge 6 commits intoModelEngine-Group:mainfrom
guotingxuan:feature/video-ops-clean
Open

feat: add video analysis operators and qwen vl service deployment files#469
guotingxuan wants to merge 6 commits intoModelEngine-Group:mainfrom
guotingxuan:feature/video-ops-clean

Conversation

@guotingxuan
Copy link
Copy Markdown

Summary

This PR adds the video analysis operator suite for DataMate, including video preprocessing, tracking, OCR, classification, summarization and event tagging.

Included

  • shared video utilities under _video_common
  • video operator implementations under runtime/ops/mapper/
  • QwenVL service deployment files under scripts/images/qwen-vl-service:
  • Dockerfile
  • start.sh
  • qwen_vl_server.py
  • technical report for implementation, models, deployment and evaluation

Notes

  • model weights are not committed to the code repository
  • open-source models and related links are documented in the technical report
  • Qwen-related capabilities are provided through a standalone HTTP service

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant