Papers with Code · CC BY-SA 4.0

ObjectNet

아카이브 평가 106건 · 마지막 적재/변경 2026-10-30

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델	원본 지표·값	논문	구현
ResNet-50 + CGC	Top-5 Accuracy 50.16 Top-1 Accuracy 31.53	Context-Gated Convolution 2019-10-12	코드 1
DeiT-L	Top-5 Accuracy 48.5 Top-1 Accuracy 31.4	Optimizing Relevance Maps of Vision Transformers Improves Robustness 2022-06-02	코드 1
Discrete ViT + Pixel	Top-1 Accuracy 30.98	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
Discrete ViT + Pyramid	Top-1 Accuracy 30.28	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
RegViT (RandAug) + Adv Pixel	Top-1 Accuracy 30.11	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
Discrete ViT	Top-1 Accuracy 29.95	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
ResNet-152	Top-5 Accuracy 49.4 Top-1 Accuracy 29.59	ObjectNet: A large-scale bias-controlled dataset for pushing the limits of object recognition models 2019-12-01	연결 없음
RegViT (RandAug) + Random Pyramid	Top-1 Accuracy 29.41	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
RegViT (RandAug)	Top-1 Accuracy 29.3	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
ResNet-50 + GroupNorm	Top-5 Accuracy 50.2 Top-1 Accuracy 29.2	Improving robustness against common corruptions by covariate shift adaptation 2020-06-30	코드 1 · 코드 2
ResNet-50 + RoHL	Top-1 Accuracy 29.2	Improving robustness against common corruptions by covariate shift adaptation 2020-06-30	코드 1 · 코드 2
RegViT (RandAug) + Random Pixel	Top-1 Accuracy 28.72	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
MLP-Mixer + Pyramid	Top-1 Accuracy 28.6	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
ResNet-50 + FixUp	Top-5 Accuracy 48.6 Top-1 Accuracy 28.5	Improving robustness against common corruptions by covariate shift adaptation 2020-06-30	코드 1 · 코드 2
ResNet-50 + MixUp (rescaled)	Top-1 Accuracy 28.37	On Mixup Regularization 2020-06-10	코드 1
DeiT-S	Top-5 Accuracy 47.3 Top-1 Accuracy 28.3	Optimizing Relevance Maps of Vision Transformers Improves Robustness 2022-06-02	코드 1
ResNet-18 + GenInt with Transfer	Top-5 Accuracy 48.02 Top-1 Accuracy 27.03	Generative Interventions for Causal Learning 2020-12-22	코드 1
MLP-Mixer	Top-1 Accuracy 25.9	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
RELICv2	Top-1 Accuracy 25.9	Pushing the limits of self-supervised ResNets: Can we outperform supervised learning without labels on ImageNet? 2022-01-13	코드 1
ViT + MixUp	Top-1 Accuracy 25.65	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
C-BYOL	Top-1 Accuracy 25.5	Compressive Visual Representations 2021-09-27	코드 1
MLP-Mixer + Pixel	Top-1 Accuracy 24.75	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
BYOL (BG_RM)	Top-1 Accuracy 23.9	Characterizing and Improving the Robustness of Self-Supervised Learning through Background Augmentations 2021-03-23	연결 없음
RELIC	Top-1 Accuracy 23.8	Pushing the limits of self-supervised ResNets: Can we outperform supervised learning without labels on ImageNet? 2022-01-13	코드 1
BYOL	Top-1 Accuracy 23	Pushing the limits of self-supervised ResNets: Can we outperform supervised learning without labels on ImageNet? 2022-01-13	코드 1
SwAV (BG_RM)	Top-1 Accuracy 21.9	Characterizing and Improving the Robustness of Self-Supervised Learning through Background Augmentations 2021-03-23	연결 없음
ViT + CutMix	Top-1 Accuracy 21.61	Pyramid Adversarial Training Improves ViT Performance 2021-11-30	코드 1
MoCo-v2 (BG_Swaps)	Top-1 Accuracy 20.8	Characterizing and Improving the Robustness of Self-Supervised Learning through Background Augmentations 2021-03-23	연결 없음
C-SimCLR	Top-1 Accuracy 20.8	Compressive Visual Representations 2021-09-27	코드 1
SeLa(v2) (reverse linear probing)	Top-5 Accuracy 48.83 Top-1 Accuracy 20.61	Measuring the Interpretability of Unsupervised Representations via Quantized Reversed Probing 2021-09-29	연결 없음