Add GroundingDINO and YOLO-World zero-shot models - Added GroundingDINO and YOLO-World for better zero-shot detection - Updated requirements.txt with ultralytics - Added visualization with distinct colors

Fix Florence-2: use base model with dense region captions and better logging

Add Florence-2 multimodal vision-language model for zero-shot detection

Improve zero-shot detection: simplified CLIP logic and better visualization

Fix zero-shot models to actually detect anomalies using local inference

Add OWL-ViT zero-shot object detection model

Add Zero Shot (Anomaly) model to app.py and remove unused api.py

Add /api/health endpoint for service monitoring and health checks

Fix: Use Gradio click handler for /api/models endpoint instead of unsupported api_endpoint decorator

Add /models API endpoint to list available detection models

Fix cv2 import scope error in rate limiting

Add rate limiting to prevent DDoS attacks

Add /api/detect endpoint for JSON detections (MonitaQC compatible)

Fix numpy array boolean ambiguity in image None check

Restructure for HuggingFace Spaces - use native Gradio API instead of FastAPI

Fix Gradio Blocks integration - attach FastAPI to Gradio app

Fix HuggingFace Spaces deployment - use Gradio Blocks and remove manual server launch

Fix port binding issue - mount Gradio at root path

Add unified FastAPI + Gradio service for MonitaQC integration

check grammer

revet

test without css

check the three columns

remove pagination

add examples

update app.py

default threshold 0.001

change the token

show case