AI Paper Atlas

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

S-JEPA: A Joint Embedding Predictive Architecture for Skeletal Action Recognition

Mohamed Abdelfattah, Alexandre Alahi

ECCV 2024 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Empirical contrastive visual learning · 12 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
update target encoder with EMA of view encoder (ablation)S-JEPA · NTU60 XSub, linear evalaccuracy (%): 1.6 -> 85.3Improved
add geometric transformations to create diverse views (ablation)S-JEPA · NTU60 XSub, linear evalaccuracy (%): 84.6 -> 85.3Improved
use cross-entropy loss over softmax distributions instead of MSE (ablation)S-JEPA · NTU60 XSub, linear evalaccuracy (%): 83.7 -> 85.3Improved
add centering of target representations (ablation)S-JEPA · NTU60 XSub, linear evalaccuracy (%): 84.5 -> 85.3Improved
mask output of target encoder instead of its input (ablation)S-JEPA · NTU60 XSub, linear evalaccuracy (%): 78.1 -> 85.3Improved
pre-train with S-JEPA before fine-tuning vs training from scratchvanilla Transformer · NTU60 XSub fine-tuningaccuracy (%): +10%Improved
pre-train with S-JEPA vs from scratch with 1% labelsvanilla Transformer · NTU60 XSub semi-supervisedaccuracy (%): +28.7 pointsImproved
predict latent representations instead of masked motion (MAMP)masked skeleton pretraining · NTU60 XSub fine-tuningaccuracy (%): 93.1 -> 93.1No clear effect
predict latent representations instead of masked motion (MAMP)masked skeleton pretraining · NTU60 XView 10% labelsaccuracy (%): 91.5 -> 91.4Hurt
predict latent representations instead of masked motion (MAMP)masked skeleton pretraining · transfer to PKU-IIaccuracy (%): 70.6 -> 71.4 (NTU60 source)Improved
set EMA update rate to 0.9999 (ablation)S-JEPA · NTU60 XSub, linear evalaccuracy (%): 83.6 (0.999) -> 85.3 (0.9999)Improved

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