Object Recognition on N-Caltech101 (Accuracy)
94.56AccuracyDSAN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DSANBackbone=R-50, SFUDA=false, Unsupervised=true2024.03 | 94.56 | — | |
| DSANS.F.=false, Unsup.=true, Backbone=R-502024.03 | 94.56 | — | |
| EventDanceBackbone=R-50, SFUDA=true2024.03 | 92.35 | 21.62 | |
| EventDanceS.F.=true, Unsup.=true, Backbone=R-502024.03 | 92.35 | — | |
| DSANBackbone=R-34, SFUDA=false, Unsupervised=true2024.03 | 89.01 | — | |
| DSANS.F.=false, Unsup.=true, Backbone=R-242024.03 | 89.01 | — | |
| EventRPG (Saliency Map)Training Method=STBP, Neural Network=Spike-VGG11, Neuron=LIF, Timesteps=10, Resolution=(128,128)2024.03 | 85.62 | — | |
| EventRPG (CAM)Training Method=STBP, Neural Network=Spike-VGG11, Neuron=LIF, Timesteps=10, Resolution=(128,128)2024.03 | 85 | — | |
| NDATraining Method=STBP-tdBN [Zheng et al., 2021], Neural Network=Spike-VGG11, Neuron=LIF, Timesteps=10, Resolution=(128,128)2024.03 | 83.7 | — | |
| SHOTBackbone=R-50, SFUDA=true2024.03 | 83.35 | 12.62 | |
| SHOT++Backbone=R-50, SFUDA=true2024.03 | 82.88 | 12.15 | |
| Ev-LaFOR + CLIPPrompt=Visual Prompt, SFUDA=false, Unsupervised=true2024.03 | 82.61 | — | |
| Ev-LaFOR + CLIPS.F.=false, Unsup.=true, Training Mode=Visual Prompt2024.03 | 82.61 | — | |
| Ev-LaFORPrompt=Text Prompt, SFUDA=false, Unsupervised=true2024.03 | 82.46 | — | |
| Ev-LaFORS.F.=false, Unsup.=true, Training Mode=Text Prompt2024.03 | 82.16 | — | |
| RPGMixModel=Spike-VGG112024.03 | 81.75 | — | |
| EventmixTraining Method=STBP, Neural Network=Pre-Act Resnet18, Neuron=PLIF, Timesteps=10, Resolution=(48, 48)2024.03 | 79.47 | — | |
| Puzzle MixModel=Spike-VGG112024.03 | 79.38 | — | |
| Zhao et al.Backbone=R-50, SFUDA=true2024.03 | 78.72 | 7.99 | |
| DSANBackbone=R-18, SFUDA=false, Unsupervised=true2024.03 | 78.45 | — | |
| DSANS.F.=false, Unsup.=true, Backbone=R-182024.03 | 78.45 | — | |
| NDATraining Method=STBP-tdBN [Zheng et al., 2021], Neural Network=Spike-VGG11, Neuron=LIF, Timesteps=10, Resolution=(48,48)2024.03 | 78.2 | — | |
| Puzzle Mix (mask only)Model=Spike-VGG112024.03 | 78.13 | — | |
| IdentityTraining Method=TET [Deng et al., 2022]2024.03 | 75.7 | — | |
| IdentityModel=Spike-VGG112024.03 | 75.7 | — | |
| EventDropTraining Method=TET [Deng et al., 2022]2024.03 | 74.04 | — | |
| EventDanceBackbone=R-34, SFUDA=true2024.03 | 72.68 | 13.9 | |
| EventDanceS.F.=true, Unsup.=true, Backbone=R-242024.03 | 72.68 | — | |
| Saliency MixModel=Spike-VGG112024.03 | 72.63 | — | |
| BaselineBackbone=R-50, SFUDA=false2024.03 | 70.73 | — | |
| EventDanceBackbone=R-18, SFUDA=true2024.03 | 66.77 | 25.96 | |
| EventDanceS.F.=true, Unsup.=true, Backbone=R-182024.03 | 66.77 | — | |
| SHOT++Backbone=R-34, SFUDA=true2024.03 | 64.54 | 5.76 | |
| SHOTBackbone=R-34, SFUDA=true2024.03 | 61.12 | 2.34 | |
| Zhao et al.Backbone=R-34, SFUDA=true2024.03 | 61.1 | 2.32 | |
| E2VIDTraining=Fine-tune, SFUDA=false, Unsupervised=false2024.03 | 59.8 | — | |
| E2VIDS.F.=false, Unsup.=true, Training Mode=Fine-tune2024.03 | 59.8 | — | |
| BaselineBackbone=R-34, SFUDA=false2024.03 | 58.78 | — | |
| FlipTraining Method=SALT [Kim & Panda, 2021a], Neural Network=Spike-VGG16, Neuron=LIF, Timesteps=20, Resolution=(80,80)2024.03 | 55 | — | |
| SHOT++Backbone=R-18, SFUDA=true2024.03 | 49.33 | 8.52 | |
| Zhao et al.Backbone=R-18, SFUDA=true2024.03 | 44.3 | 3.49 | |
| SHOTBackbone=R-18, SFUDA=true2024.03 | 44.26 | 3.45 | |
| Wang et al.SFUDA=false, Unsupervised=true2024.03 | 43.5 | — | |
| Wang et al.S.F.=false, Unsup.=true2024.03 | 43.5 | — | |
| Wang et al.SFUDA=true, Unsupervised=true2024.03 | 42.7 | — | |
| Wang et al.S.F.=true, Unsup.=true2024.03 | 42.7 | — | |
| BaselineBackbone=R-18, SFUDA=false2024.03 | 40.81 | — | |
| Wang et al. + CLIPSFUDA=true, Unsupervised=false2024.03 | 39.7 | — | |
| Wang et al. + CLIPS.F.=true, Unsup.=false2024.03 | 39.7 | — | |
| E2VID + CLIPTraining=Scratch, SFUDA=false, Unsupervised=false2024.03 | 9.4 | — | |
| E2VID + CLIPS.F.=false, Unsup.=true, Training Mode=Scratch2024.03 | 9.4 | — |