notesum.ai
Published at December 3Composing Open-domain Vision with RAG for Ocean Monitoring and Conservation
cs.CV
cs.LG
Released Date: December 3, 2024
Authors: Sepand Dyanatkar1, Angran Li1, Alexander Dungate1
Aff.: 1OnDeck Fisheries AI

| Accuracy | |||
|---|---|---|---|
| Method | Top-1 | Top-2 | Top-3 |
| InceptionV3 (Baseline) | 0.7501 | 0.8312 | 0.9408 |
| VLM-RAG (Ours, Final Prediction) | 0.8403 | N/A (single answer) | N/A (single answer) |
| VLM-RAG (Ours, RAG Retrieval) | 0.8684 | 0.9527 | 0.9781 |