ImageNet-trained CNNs are not biased towards texture: Revisiting feature reliance through controlled suppression

Documenti analoghi

• Rank-based Geographical Regularization: Revisiting Contrastive Self-Supervised Learning for Multispectral Remote Sensing Imagery
  di: Burgert, Tom, et al.
  Pubblicazione: (2026)
• Estimating Physical Information Consistency of Channel Data Augmentation for Remote Sensing Images
  di: Burgert, Tom, et al.
  Pubblicazione: (2024)
• Noise-Adaptive Regularization for Robust Multi-Label Remote Sensing Image Classification
  di: Burgert, Tom, et al.
  Pubblicazione: (2026)
• On the Effectiveness of Methods and Metrics for Explainable AI in Remote Sensing Image Scene Classification
  di: Klotz, Jonas, et al.
  Pubblicazione: (2025)
• CSMoE: An Efficient Remote Sensing Foundation Model with Soft Mixture-of-Experts
  di: Hackel, Leonard, et al.
  Pubblicazione: (2025)