Object Classification on N-CARS (test)
96.8AccuracyMVF-Net
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MVF-NetPre-training=Pretrained on ImageNet [9], Backbone=ResNet-34 [21]2021.06 | 96.8 | — | |
| GETInput Type=Token, Params (M)=4.52023.10 | 96.7 | — | |
| TEFormerSize=2-256, Step=16, Batch-Size=322026.01 | 95.95 | — | |
| M-LSTMPre-training=Pretrained on ImageNet [9], Backbone=ResNet-34 [21]2021.06 | 95.7 | — | |
| SpikformerSize=2-256, Step=16, Batch-Size=322026.01 | 95.6 | — | |
| EV-VGCNNPre-training=Without pretraining2021.06 | 95.3 | — | |
| EV-VGCNN (Ours)2021.06 | 95.3 | — | |
| EV-VGCNNInput Type=Voxel, Params (M)=0.82023.10 | 95.3 | — | |
| QKFormerSize=2-256, Step=16, Batch-Size=322026.01 | 95.29 | — | |
| SSCRepresentation=Event-Histogram, Async=false2022.03 | 94.5 | 321 | |
| AEGNNRepresentation=Graph, Async=true2022.03 | 94.5 | 0.47 | |
| AsyNetPre-training=Without pretraining2021.06 | 94.4 | — | |
| EV-VGCNN (Ours)Variation=w/ SFRL2021.06 | 94.4 | — | |
| AsyNetRepresentation=Event-Histogram, Async=true2022.03 | 94.4 | 21.5 | |
| AMAEInput Type=Frame, Params (M)=21.82023.10 | 93.6 | — | |
| Nested-TInput Type=Token, Params (M)=4.22023.10 | 93.3 | — | |
| VMV-GCNInput Type=Voxel, Params (M)=0.92023.10 | 93.2 | — | |
| EvS-SRepresentation=Graph, Async=true2022.03 | 93.1 | 6.1 | |
| EvSInput Type=Graph2023.10 | 93.1 | — | |
| Swin-T v2Input Type=Token, Params (M)=6.92023.10 | 92.8 | — | |
| M-LSTMPre-training=Without pretraining, Backbone=ResNet-34 [21]2021.06 | 92.7 | — | |
| MVF-NetPre-training=Without pretraining, Backbone=ResNet-34 [21]2021.06 | 92.7 | — | |
| YOLERepresentation=Event-Histogram, Async=true2022.03 | 92.7 | 328.16 | |
| MVF-NetInput Type=Frame, Params (M)=33.52023.10 | 92.7 | — | |
| ESTMeasurement=time stamps, Kernel=learnt2019.04 | 92.5 | — | |
| ESTPre-training=Pretrained on ImageNet [9], Backbone=ResNet-34 [21]2021.06 | 92.5 | — | |
| ESTRepresentation=Event-Histogram, Async=false2022.03 | 92.5 | 1,050 | |
| ESTPre-training=Without pretraining, Backbone=ResNet-34 [21]2021.06 | 91.9 | — | |
| ESTMeasurement=time stamps, Kernel=trilinear2019.04 | 91.7 | — | |
| NVS-SRepresentation=Graph, Async=true2022.03 | 91.5 | 5.2 | |
| RG-CNNs2021.06 | 91.4 | — | |
| RG-CNNsInput Type=Voxel, Params (M)=19.52023.10 | 91.4 | — | |
| E2VIDEvaluation Protocol=fine-tuned, Backbone/Classifier=ResNet-182019.06 | 91 | — | |
| HATS + ResNet-342019.04 | 90.9 | — | |
| PointNet++Input representation=Proposed representation2021.06 | 90.7 | — | |
| HATSBackbone/Classifier=ResNet-182019.06 | 90.4 | — | |
| HATSBackbone/Classifier=linear SVM2019.06 | 90.2 | — | |
| HATS2019.04 | 90.2 | — | |
| HATS2021.06 | 90.2 | — | |
| HATSRepresentation=Time-Surface, Async=true2022.03 | 90.2 | 0.03 | |
| HATSInput Type=Frame2023.10 | 90.2 | — | |
| Voxel GridMeasurement=polarity, Kernel=trilinear2019.04 | 86.5 | — | |
| Two-Channel ImageMeasurement=count, Kernel=trilinear2019.04 | 86.1 | — | |
| PointNet++Input representation=Point-wise2021.06 | 80.9 | — | |
| Gabor-SNN2019.04 | 78.9 | — | |
| EventNet2021.06 | 75 | — | |
| HOTS2019.06 | 62.4 | — | |
| HOTS2019.04 | 62.4 | — | |
| HOTS2021.06 | 62.4 | — | |
| HOTSRepresentation=Time-Surface, Async=true2022.03 | 62.4 | 14 | |
| H-First2019.04 | 56.1 | — | |
| H-First2021.06 | 56.1 | — | |
| H-FirstRepresentation=Spike, Async=true2022.03 | 56.1 | — |