- Live Classifier: Train custom classes with labeled examples, test classification
- Few-Shot Learning: 3-class system (A/B/C) with drag-drop training
- Incremental Learning: Positive/negative examples with prototype visualization
- Feedback Learning: Track predictions and accuracy over time
- Memory Bank: View stored embeddings, export/import as JSON
- Camera Training: Train using webcam with single/auto-capture modes
All demos use real CNN embeddings (512-dim) with prototypical networks
for classification. Includes cosine similarity scoring and confidence bars.
Co-Authored-By: claude-flow <ruv@ruv.net>
- Wait for video to be fully ready before processing
- Add camera warmup delay
- Increase sensitivity (10x amplification, lower threshold)
- Center crop video for consistent detection
- Add better error handling and logging
Co-Authored-By: claude-flow <ruv@ruv.net>
Interactive Demos:
- Similarity Search: Click image to find similar ones
- Motion Detection: Real-time scene change via embeddings
- A/B Comparison: Side-by-side image comparison
- Batch Processor: Process up to 20 images at once
- Embedding Explorer: Visualize 512-dim features with stats
- Anomaly Detection: Find outlier images in a set
Self-Learning Code Examples:
- Self-Learning System: Adaptive classifier with feedback loop
- Incremental Learning: Online learning without forgetting
- Few-Shot Learning: Learn from 1-5 examples per class
- Image Retrieval: Searchable image database
- Content Moderation: Flag similar inappropriate content
- Visual Recommendations: Recommend similar items
- Video Keyframes: Extract unique frames
Co-Authored-By: claude-flow <ruv@ruv.net>
- Real-time embedding extraction at ~30+ FPS
- Live FPS and latency display
- Reference image comparison with similarity score
- Live embedding visualization
- Center-crop to square for consistent input
Co-Authored-By: claude-flow <ruv@ruv.net>
- Use object format for init: { module_or_path: url }
- Use WasmCnnEmbedder (not CnnEmbedder)
- Use snake_case methods: embedding_dim, cosine_similarity
Co-Authored-By: claude-flow <ruv@ruv.net>
