Visual Question Answering on PartNet-Reasoning Chair 1.0 (test)
85.4Exist Part Accuracy3D-CG
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| 3D-CGArchitecture=Neural-Symbolic2022.07 | 85.4 | 68.7 | 92.2 | |
| BAE+LArchitecture=Neural-Symbolic2022.07 | 72.4 | 70.5 | 68 | |
| CVX+LArchitecture=Neural-Symbolic2022.07 | 71.9 | 72.3 | 48.8 | |
| MACArchitecture=Neural Network2022.07 | 65.2 | 53.9 | 78.1 | |
| NDF+LSTMArchitecture=Neural Network2022.07 | 55.7 | 54.2 | 71.6 | |
| PointNet+LSTMArchitecture=Neural Network2022.07 | 52.3 | 41.6 | 63.4 |