Cross-modal sequential scene retrieval on 3RScan 10 candidates (test)
89Rt@1SG2Loc
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SG2Loc# of frames=50, Inference Time (ms)=15.0, Storage (MB)=5.42026.06 | 89 | 96 | 98 | |
| SG2Loc# of frames=25, Inference Time (ms)=7.5, Storage (MB)=5.42026.06 | 88 | 96 | 98 | |
| AnyLocInference Time (ms)=1826.4, Storage (MB)=5720.32026.06 | 88 | 95 | 98 | |
| SG2Loc# of frames=10, Inference Time (ms)=3.0, Storage (MB)=5.42026.06 | 86 | 95 | 98 | |
| SG2Loc# of frames=5, Inference Time (ms)=1.5, Storage (MB)=5.42026.06 | 84 | 95 | 98 | |
| SceneGraphLoc# of frames=1, Inference Time (ms)=0.3, Storage (MB)=5.42026.06 | 82 | 94 | 98 | |
| CVNetInference Time (ms)=60.0, Storage (MB)=239.12026.06 | 79 | 91 | 95 |