| Dataset |
Model |
SNN |
Acc(%) |
| PAF |
HMAX SNN (Xiao et al., 2019)
|
✔ |
55.00 |
| STCA (Gu et al., 2019)
|
✔ |
71.20 |
| Motion SNN (Liu et al., 2021b)
|
✔ |
78.10 |
| MST (Wang et al., 2023)
|
✔ |
88.21 |
| Swin-T (BN) (Wang et al., 2023)
|
✔ |
90.14 |
| EV-ACT (Gao et al., 2023)
|
✗ |
92.60 |
| ExACT (Zhou et al., 2024)
|
✗ |
94.83 |
| SpikMamba(Ours) |
✔ |
96.28 |
| HARDVS |
X3D (Feichtenhofer, 2020)
|
✗ |
45.82 |
| SlowFast (Feichtenhofer et al., 2019)
|
✗ |
46.54 |
| ACTION-Net (Wang et al., 2021)
|
✗ |
46.85 |
| R2Plus1D (Tran et al., 2018)
|
✗ |
49.06 |
| ResNet18 (He et al., 2016)
|
✗ |
49.20 |
| TAM (Liu et al., 2021a)
|
✗ |
50.41 |
| C3D (Tran et al., 2015)
|
✗ |
50.52 |
| ESTF (Wang et al., 2024b)
|
✗ |
51.22 |
| Video-SwinTrans (Liu et al., 2022)
|
✗ |
51.91 |
| TSM (Lin et al., 2019)
|
✗ |
52.63 |
| ExACT (Zhou et al., 2024)
|
✗ |
90.10 |
| SpikMamba(Ours) |
✔ |
97.32 |
| DVSGesture |
Time-surfaces (Maro et al., 2020)
|
✔ |
90.62 |
| SNN eRBP (Kaiser et al., 2019)
|
✔ |
92.70 |
| Slayer (Shrestha and Orchard, 2018)
|
✔ |
93.64 |
| DECOLLE (Kaiser et al., 2020)
|
✔ |
95.54 |
| EvT (Sabater et al., 2022)
|
✔ |
96.20 |
| TBR (Innocenti et al., 2021)
|
✗ |
97.73 |
| EventTransAct (de Blegiers et al., 2023)
|
✗ |
97.92 |
| ExACT (Zhou et al., 2024)
|
✗ |
98.86 |
| SpikMamba(Ours) |
✔ |
99.01 |
| E-FAction |
CLIP-L (Radford et al., 2021)
|
✗ |
61.90 |
| ResNet3D-N (Kim et al., 2021)
|
✗ |
65.60 |
| ResNet3D-K (Kay et al., 2017)
|
✗ |
66.30 |
| MASTAF (Liu et al., 2023)
|
✗ |
67.10 |
| ExACT (Zhou et al., 2024)
|
✗ |
67.93 |
| SpikMamba(Ours) |
✔ |
71.02 |
