Object Classification on N-Caltech101 (test)
90.2AccuracyEvDistill
Evaluation results
| Method | Links | ||
|---|---|---|---|
| EvDistillTraining data=Real events, Use GT labels=false, Mode=fine-tune2021.11 | 90.2 | — | |
| EvDistillTraining data=Real events, Use GT labels=false, Accumulated events=20K2021.11 | 89.6 | — | |
| MVF-NetPre-training=Pretrained on ImageNet [9], Backbone=ResNet-34 [21]2021.06 | 87.1 | — | |
| E2VIDEvaluation Protocol=fine-tuned, Backbone/Classifier=ResNet-182019.06 | 86.6 | — | |
| E2VIDTraining data=Intensity images, Use GT labels=true2021.11 | 86.6 | — | |
| E2VIDAsync.=Synchronous2026.06 | 86.6 | — | |
| TokSwinTAsync.=A2S2026.06 | 86.5 | — | |
| EVA-LAsync.=A2S2026.06 | 86.3 | — | |
| M-LSTMPre-training=Pretrained on ImageNet [9], Backbone=ResNet-34 [21]2021.06 | 85.7 | — | |
| PillarAsync.=Synchronous2026.06 | 85.3 | — | |
| Matrix-LSTMAsync.=Synchronous2026.06 | 84.3 | — | |
| ESTPre-training=Pretrained on ImageNet [9], Backbone=ResNet-34 [21]2021.06 | 83.7 | — | |
| ESTAsync.=Synchronous2026.06 | 83.7 | — | |
| E2VIDEvaluation Protocol=transfer learning2019.06 | 82.1 | — | |
| ESTTraining data=Real events, Use GT labels=true2021.11 | 81.7 | — | |
| ESTRepresentation=Event-Histogram, Async=false2022.03 | 81.7 | 4,150 | |
| VID2ETraining data=Synthetic events, Use GT labels=true2021.11 | 80.7 | — | |
| CogniSNNNetwork=ER-RGA-7, Param(M)=1.51, T=52025.12 | 80.64 | — | |
| CogniSNNNetwork=WS-RGA-7, Param(M)=1.51, T=82025.12 | 80.23 | — | |
| CogniSNNNetwork=ER-RGA-7, Param(M)=1.51, T=82025.12 | 79.32 | — | |
| SSNNNetwork=VGG-9, T=82025.12 | 79.25 | — | |
| TIMNetwork=Spikformer, T=102025.12 | 79 | — | |
| CogniSNNNetwork=WS-RGA-7, Param(M)=1.51, T=52025.12 | 78.62 | — | |
| NDANetwork=VGG11, T=102025.12 | 78.2 | — | |
| SSNNNetwork=VGG-9, T=52025.12 | 77.97 | — | |
| TT-SNNNetwork=ResNet34, T=62025.12 | 77.8 | — | |
| SSCRepresentation=Event-Histogram, Async=false2022.03 | 76.1 | 1,621 | |
| EvS-SRepresentation=Graph, Async=true2022.03 | 76.1 | 11.5 | |
| ESTPre-training=Without pretraining, Backbone=ResNet-34 [21]2021.06 | 75.3 | — | |
| EV-VGCNNPre-training=Without pretraining2021.06 | 74.8 | — | |
| EV-VGCNN (Ours)2021.06 | 74.8 | — | |
| AsyNetPre-training=Without pretraining2021.06 | 74.5 | — | |
| AsyNetRepresentation=Event-Histogram, Async=true2022.03 | 74.5 | 202 | |
| AsynetAsync.=Asynchronous2026.06 | 74.5 | — | |
| M-LSTMPre-training=Without pretraining, Backbone=ResNet-34 [21]2021.06 | 73.8 | — | |
| EV-VGCNN (Ours)Variation=w/ SFRL2021.06 | 73.7 | — | |
| SpikformerNetwork=Spikformer, Param(M)=2.57, T=52025.12 | 72.8 | — | |
| YOLERepresentation=Event-Histogram, Async=true2022.03 | 70.2 | 3,659 | |
| HATSBackbone/Classifier=ResNet-182019.06 | 70 | — | |
| HATS-ResNet34Training data=Real events, Use GT labels=true, Backbone=ResNet342021.11 | 69.1 | — | |
| MVF-NetPre-training=Without pretraining, Backbone=ResNet-34 [21]2021.06 | 68.7 | — | |
| FARSE-CNNAsync.=Asynchronous2026.06 | 68.7 | — | |
| NVS-SRepresentation=Graph, Async=true2022.03 | 67 | 7.8 | |
| NVS-SAsync.=Asynchronous2026.06 | 67 | — | |
| AEGNNRepresentation=Graph, Async=true2022.03 | 66.8 | 7.31 | |
| AEGNNAsync.=Asynchronous2026.06 | 66.8 | — | |
| DARTRepresentation=Time-Surface, Async=true2022.03 | 66.4 | — | |
| RG-CNNTraining data=Real events, Use GT labels=true2021.11 | 65.7 | — | |
| RG-CNNs2021.06 | 65.7 | — | |
| HATSBackbone/Classifier=linear SVM2019.06 | 64.2 | — | |
| HATSTraining data=Real events, Use GT labels=true2021.11 | 64.2 | — | |
| HATS2021.06 | 64.2 | — | |
| HATSRepresentation=Time-Surface, Async=true2022.03 | 64.2 | 4.3 | |
| PointNet++Input representation=Proposed representation2021.06 | 62.1 | — | |
| PointNet++Input representation=Point-wise2021.06 | 50.3 | — | |
| EventNet2021.06 | 42.5 | — | |
| HOTS2019.06 | 21 | — | |
| HOTS2021.06 | 21 | — | |
| HOTSRepresentation=Time-Surface, Async=true2022.03 | 21 | 54 | |
| HOTSAsync.=Asynchronous2026.06 | 21 | — | |
| H-First2021.06 | 5.4 | — | |
| H-FirstRepresentation=Spike, Async=true2022.03 | 5.4 | — |