Video-to-Text Retrieval on Video-QA
76.8Recall@10SPI-Multimodal
Evaluation results
| Method | Links | |
|---|---|---|
| SPI-MultimodalLatency (ms)=1252025.11 | 76.8 | |
| Video-RAGLatency (ms)=1652025.11 | 74.1 | |
| ImageBind-RAGLatency (ms)=1702025.11 | 70.1 | |
| MMRetrieverLatency (ms)=1452025.11 | 68.7 | |
| FrozenBiLMLatency (ms)=1552025.11 | 66.2 | |
| BLIP-2Latency (ms)=1802025.11 | 61.4 |