AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Revisiting a kNN-based Image Classification System with High-capacity Storage

Kengo Nakata, Youyang Ng, Daisuke Miyashita, Asuka Maki, Yu-Chieh Lin, Jun Deguchi

ECCV 2022 · Architectures and efficiency › Pruning and efficient tuning · approach: Efficient ViT adaptation · 24 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
store training features and labels and use kNN instead of zero-shot classifierCLIP ViT-L/14 · ImageNet-1ktop-1 accuracy: 75.3 -> 79.7Improved
store training features and labels and use kNN instead of zero-shot classifierCLIP ResNet-50 · CIFAR-100top-1 accuracy: 41.6 -> 55.7Improved
store training features and labels and use kNN instead of zero-shot classifierCLIP ViT-L/14 · STL-10accuracy: 99.3 -> 99.6No clear effect
use CLIP-pretrained encoder instead of ImageNet-supervised encoder (ablation)kNN system · CIFAR-100accuracy: 64.7 -> 74.3Improved
use CLIP-pretrained encoder instead of MAE encoder (ablation)kNN system · ImageNet-1ktop-1 accuracy: 26.7 -> 74.0Improved
use ImageNet-supervised encoder instead of CLIP on its own training set (ablation)kNN system · ImageNet-1ktop-1 accuracy: 74.0 -> 74.9Mixed
increase number of stored pairs (12.8k to 1.28M) (ablation)kNN system · ImageNet-1ktop-1 accuracyImproved
use larger encoder model (ablation)kNN system · ImageNet-1ktop-1 accuracy: 65.0 -> 79.7Improved
add horizontal-flip augmentation of support set (ablation)kNN system · CIFAR-100 and ImageNet-1ktop-1 accuracy: +1.1 for CIFAR-100 ResNet-50 and +0.5 for ImageNet-1k ResNet-101Improved
add horizontal-flip augmentation of support set (ablation)kNN system · CIFAR-10 and STL-10accuracy: -0.1 to +0.8No clear effect
use kNN over stored features instead of fine-tuning methods in task-incremental learningkNN system · Split CIFAR-100 (ResNet-50)average accuracy: 80.9 -> 90.8Improved
use kNN over stored features instead of fine-tuning methods in class-incremental learningkNN system · Split CIFAR-100 (ResNet-50)average accuracy: 56.6 -> 67.9Improved

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