CAPE-JEPA - Cross-modal Adversarial Probabilistic Embedding JEPA

概要

Cross-modal remote sensing image retrieval (RS-CMIR) requires representations that are simultaneously modality-invariant and semantically discriminative, despite the profound physical gap between heterogeneous sensors such as Synthetic Aperture Radar (SAR) and multispectral imagery. Recent JEPA-based Predictive self-supervised approaches such as X-JEPA formulate cross-modal alignment as a deterministic latent forecasting task, implicitly assuming uniform patch predictability and leaving cross-modal distributional alignment unaddressed. We propose CAPE-JEPA, a probabilistic cross-modal Joint Embedding Predictive Architecture with three contributions: (i) a cross-attentional Gaussian predictor outputting per-patch parameters (mu, sigma^2) conditioned on opposite-modality context, trained via negative log-likelihood to model inherent cross-sensor prediction uncertainty; (ii) adversarial distributional regularization combining unimodal isotropic Gaussian regularization with a novel cross-modal adversarial variant that aligns modality distributions via worst-case Epps-Pulley characteristic function tests without explicit sample pairing; and (iii) in-training contrastive learning with attention-pooled heads for scene-level discriminative grounding. Evaluated on BigEarthNet-14K (BEN-14K) and fMoW across SAR, multispectral, and RGB modalities, CAPE-JEPA surpasses X-JEPA by up to 13.74 points on BEN-14K and 15.59 points on fMoW, outperforming all baselines including SkySense and CROMA across both unimodal and cross-modal directions while using fewer parameters (407.30M vs. 432.72M).

参考リンク

• Paper ↗

リンク先のライセンス・再現性・データ条件をご確認ください。自動分類や実装の目安は保証ではありません。

実装を始めるプロンプト

参考リンクを確認したうえで、次の内容をコピーして利用できます。

# 論文・記事
CAPE-JEPA - Cross-modal Adversarial Probabilistic Embedding JEPA

# 概要
Cross-modal remote sensing image retrieval (RS-CMIR) requires representations that are simultaneously modality-invariant and semantically discriminative, despite the profound physical gap between…

# 原典
https://arxiv.org/abs/2611.06247

上記の手法を実装してください。実行前に依存関係とライセンスを確認してください。