AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Event Camera Data Pre-training

Yan Yang, Liyuan Pan, Liu Liu

ICCV 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 44 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use event-specific data augmentations applied before event image conversion (ablation)Event pre-training (ViT-S/16) · N-ImageNet linear probing, 100 epochstop-1 accuracy: 46.1% -> 53.5%Improved
scale backbone from ViT-S/16 to ViT-L/16 (ablation)Event pre-training · N-ImageNet fine-tuningtop-1 accuracy: 64.84 -> 71.05Improved
use proposed method instead of MAE pre-training on event dataViT backbones · N-ImageNet fine-tuningtop-1 accuracy: 55.45 -> 64.83Improved
pre-train with proposed framework vs training ViT-S/16 from scratchViT-S/16 · N-ImageNettop-1 accuracy: 46.70 -> 64.83Improved
apply RGB-image-image SSL methods (iBoT, BeiT, MoCo-v3) directly to event dataExisting SSL methods · N-ImageNet fine-tuningtop-1 accuracy: iBoT 19.55; MoCo-v3 45.77 vs ViT-S/16 scratch 46.70Hurt
pre-train with proposed framework on semantic segmentation (ResNet50) instead of ViT-S/16Event pre-training · DSECmIoU: 47.91 -> 59.16Improved
pre-train ResNet50 with proposed method vs training from scratchResNet50 · DSECmIoU: 57.60 -> 59.16Improved
pre-train ViT-S/16 with proposed method vs scratchViT-S/16 · MVSEC indoor_flying1AEE: 0.68 -> 0.61Improved
pre-train ViT-S/16 with proposed method vs scratchViT-S/16 · DDD17mIoU: 48.76 -> 54.66Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy