Audio-visual Zero-Shot Classification on VGGSound GZSL (test)
29.96S ScoreDEVISE
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| DEVISEMethod type=ZSL2022.03 | 29.96 | 1.94 | 3.64 | 4.72 | — | |
| DEVISEVenue=NeurIPS’132023.11 | 29.96 | 1.94 | 3.64 | 4.72 | — | |
| ALEMethod type=ZSL2022.03 | 26.13 | 1.72 | 3.23 | 4.97 | — | |
| AVGZSLNetFeature type=main2023.08 | 18.05 | 3.48 | 5.83 | 5.28 | — | |
| SJEMethod type=ZSL2022.03 | 16.94 | 2.72 | 4.69 | 3.22 | — | |
| SJEVenue=CVPR’202023.11 | 16.94 | 2.72 | 4.69 | 3.22 | — | |
| Hyper-multipleFeature Type=cls feature, Number of adaptive curvatures (Ne)=32023.08 | 15.62 | 6 | 8.67 | 7.31 | — | |
| Hyper-multipleVenue=ICCV’232023.11 | 15.62 | 6 | 8.67 | 7.31 | — | |
| AVGZSLNet2022.07 | 15.02 | 3.19 | 5.26 | 4.81 | — | |
| AVGZSLNetFeature Type=cls feature2023.08 | 15.02 | 3.19 | 5.26 | 4.81 | — | |
| Hyper-multipleFeature type=main, Number of adaptive curvatures (Ne)=22023.08 | 15.02 | 6.75 | 9.32 | 7.97 | — | |
| AVGZSLNetMethod type=Audio-visual ZSL2022.03 | 15.02 | 3.19 | 5.26 | 4.81 | — | |
| AVGZSLNetVenue=WACV’212023.11 | 15.02 | 3.19 | 5.26 | 4.81 | — | |
| AVCAFeature type=main2023.08 | 14.9 | 4 | 6.31 | 6 | — | |
| KDAVenue=Ours2023.11 | 14.64 | 6.63 | 9.12 | 8 | — | |
| Attention Fusion2022.07 | 14.13 | 3 | 4.95 | 3.37 | — | |
| Perceiver2022.07 | 13.25 | 3.03 | 4.93 | 3.44 | — | |
| Hyper-alignmentFeature type=main, Curvature (c)=-0.52023.08 | 13.22 | 5.01 | 7.27 | 6.14 | — | |
| ACFSVenue=IJCNN’232023.11 | 12.87 | 5.22 | 7.43 | 6.03 | — | |
| AVCA2022.07 | 12.63 | 6.19 | 8.31 | 6.91 | — | |
| TCAF2022.07 | 12.63 | 6.72 | 8.77 | 7.41 | — | |
| TCAFInput features=audio-visual2022.07 | 12.63 | 6.72 | 8.77 | 7.41 | — | |
| AVCAFeature Type=cls feature2023.08 | 12.63 | 6.19 | 8.31 | 6.91 | — | |
| AVCAMethod type=Audio-visual ZSL2022.03 | 12.63 | 6.19 | 8.31 | 6.91 | — | |
| AVCAVenue=CVPR’222023.11 | 12.63 | 6.19 | 8.31 | 6.91 | — | |
| TCaFVenue=ECCV’222023.11 | 12.63 | 6.72 | 8.77 | 7.41 | — | |
| Hyper-singleFeature Type=cls feature2023.08 | 12.56 | 5.03 | 7.18 | 5.47 | — | |
| Hyper-alignmentFeature Type=cls feature, Curvature (c)=-0.12023.08 | 12.5 | 6.44 | 8.5 | 7.25 | — | |
| CJME2022.07 | 10.86 | 2.22 | 3.68 | 3.72 | — | |
| CJMEFeature Type=cls feature2023.08 | 10.86 | 2.22 | 3.68 | 3.72 | — | |
| CJMEMethod type=Audio-visual ZSL2022.03 | 10.86 | 2.22 | 3.68 | 3.72 | — | |
| CJMEVenue=WACV’202023.11 | 10.86 | 2.22 | 3.68 | 3.72 | — | |
| Hyper-singleFeature type=main2023.08 | 9.79 | 6.23 | 7.62 | 6.46 | — | |
| CJMEFeature type=main2023.08 | 8.69 | 4.78 | 6.17 | 5.16 | — | |
| APNMethod type=ZSL2022.03 | 6.46 | 6.13 | 6.29 | 6.5 | — | |
| APNVenue=IJCV’222023.11 | 6.46 | 6.13 | 6.29 | 6.5 | — | |
| TCAF-audioInput features=audio2022.07 | 5.11 | 4.06 | 4.53 | 4.28 | — | |
| TCAF-visualInput features=visual2022.07 | 3.97 | 3.12 | 3.5 | 3.19 | — |