AI Paper Atlas

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

Towards Latent Masked Image Modeling for Self-Supervised Visual Representation Learning

Yibing Wei, Abhinav Gupta, Pedro Morgado

ECCV 2024 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Masked autoencoder pretraining · 17 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 momentum target encoder instead of joint optimization to prevent collapse (ablation)Latent MIM · IN100NN accuracy: 0.9 -> 18.8Improved
replace MSE reconstruction with InfoNCE patch discrimination (ablation)Latent MIM · IN100NN accuracy: 18.8 -> 30.0Improved
L1 loss instead of MSE for latent reconstruction (ablation)Latent MIM · IN100NN accuracy: 18.8 -> 15.2Hurt
raise mask ratio from 0.75 to 0.9 (ablation)Latent MIM · IN100NN accuracy: 30.0 -> 53.3Improved
add non-contiguous stochastic masking with gap 4 (ablation)Latent MIM · IN100NN accuracy: 53.3 -> 54.6Improved
add patch similarity constraint regularizer (ablation)Latent MIM · IN100NN accuracy: 54.6 -> 56.8Improved
switch from self-attention to cross-attention decoder with visual cues (ablation)Latent MIM · IN100NN accuracy: 59.7 -> 63.3Improved
add MLP projector before decoder (ablation)Latent MIM self-attention decoder · IN100NN accuracy: 56.8 -> 59.7Improved
increase decoder depth from 3 to 8 layers (ablation)Latent MIM · IN100NN accuracy: 63.3 -> 35.9Hurt
latent MIM instead of pixel MIM (MAE) on linear probingViT-B/16 ImageNet-1k · ImageNet-1kLP accuracy: 67.8 -> 72.0Improved
latent MIM instead of pixel MIM on IN100 fine-tuning (before decoder fixes) (ablation)Latent MIM variants · IN100fine-tuning accuracy: 79.7 (momentum) vs 88.3 (MAE); 89.9 finalMixed

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