Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 5, 2025Complex & Intelligent SystemsOpen Access

MDFAC: multi-dimensional feature adaptive calibration for generalized few-shot object detection

View Full Paper
Ask AI
Bookmark
Share

Authors

KXKailin XieJLJinxiang LaiZSZhen-Su She

Discussion

Loading...

Member takes

Overview

Proposed MDFAC enhances detection performance in few-shot object detection, indicating reduced catastrophic forgetting.

Key Points

  • MDFAC improves average detection performance for novel classes by 1.8% on nAP50 and 2.1% on nAP.
  • The method effectively manages feature distribution shifts while minimizing catastrophic forgetting in base classes.
  • MDFAC integrates the Equiangular Tight Frame Guidance Module to maintain base class knowledge during training.
  • Adaptive calibration classification dynamically adjusts attention based on real-time detection frequency.

Cite This Study

Xie et al. (2025) studied this question.

synapsesocial.com/papers/68bb421a2b87ece8dc9585echttps://doi.org/10.1007/s40747-025-02053-x
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Fine-Grained Prototypes Distillation for Few-Shot Object Detection2024 · 68 citations
  2. 2Semantic Enhanced Few-shot Object Detection2024
  3. 3HALD-FSOD: Hierarchical Adaptive Learning with Asymmetric Margin and Loss-Aware Dynamic Weighting for Few-Shot Object Detection in Remote Sensing Imagery2026
  4. 4BM-FSOD: Few-Shot Object Detection Method Based on Background Reconstruction and Multi-Channel Interactive Feature Fusion2026
  5. 5FSNA: Few-Shot Object Detection via Neighborhood Information Adaption and All Attention2024 · 5 citations