ICLR2020 cross-domain few-shot classification via learned feature-wise transformation
跨域小樣本分類 paper code 摘要 摘要：小樣本分類旨在識別每個類別中只有少數標記圖像的新類別。現有的基於度量的小樣本分類算法通過使用學習度量函數將查詢圖像的特徵嵌入與少數標記圖像(支持示例)的特徵嵌入進行比較來預測類別。雖然已經證明了這些方法有很好的性能，但是由於域之間的特徵分佈存在很大的差異，這些方法往往不能推廣到不可見的域。在這項工作中，我們解決了基於度量的方法在領域轉移下的少樣本分
相關文章
- 1. 【論文閱讀】[meta learning]cross-domain few-shot classification via learned feature-wise transformation.
- 2. Cross-Domain Few-Shot Classification VIA Learned Feature-Wise Transformation 論文解讀
- 3. [AAAI2018]Learning Structured Representation Representation for Text Classification via Reinforcemen
- 4. Unsupervised Person Re-identification via Multi-label Classification
- 5. nonlinear transformation
- 6. Text Classification
- 7. 論文閱讀筆記《MetAdapt: Meta-Learned Task-Adaptive Architecture for Few-Shot Classification》
- 8. 【圖像篡改 ECCV2018】Fighting Fake News: Image Splice Detection via Learned Self-Consistency
- 9. 論文筆記-Depth Estimation via Affinity Learned with Convolutional Spatial Propagation Network
- 10. Relation Classification via Convolutional Deep Neural Network 論文筆記
- 更多相關文章...
- • PHP is_uploaded_file() 函數 - PHP參考手冊
- • R 包 - R 語言教程
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. 【論文閱讀】[meta learning]cross-domain few-shot classification via learned feature-wise transformation.
- 2. Cross-Domain Few-Shot Classification VIA Learned Feature-Wise Transformation 論文解讀
- 3. [AAAI2018]Learning Structured Representation Representation for Text Classification via Reinforcemen
- 4. Unsupervised Person Re-identification via Multi-label Classification
- 5. nonlinear transformation
- 6. Text Classification
- 7. 論文閱讀筆記《MetAdapt: Meta-Learned Task-Adaptive Architecture for Few-Shot Classification》
- 8. 【圖像篡改 ECCV2018】Fighting Fake News: Image Splice Detection via Learned Self-Consistency
- 9. 論文筆記-Depth Estimation via Affinity Learned with Convolutional Spatial Propagation Network
- 10. Relation Classification via Convolutional Deep Neural Network 論文筆記