3DMV-VQA
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| 3D Visual Question Answering | 3DMV-VQA (held-out) | Concept Accuracy68.9 | 10 | |
| 3D Visual Question Answering | 3DMV-VQA single room only subset of 1,212 scenes | Concept Accuracy70.2 | 2 |
Showing 2 of 2 rows