[ICCV 2017] Attributes2Classname: A Discriminative Model for Attribute-Based Unsupervised Zero-ShotL
[paper] [code] 動機： 無監督ZSL指的是測試類沒有屬性標註。直接建立視覺特徵和類別名稱詞嵌入的關聯很難。類別名稱詞嵌入包含更多非視覺信息，而視覺特徵主要包含視覺信息。通過屬性來連接，屬性包含了圖像域的視覺線索，同時是語言域的一個項，將其作爲中間表示。 方法： 建立屬性組合的詞嵌入和類別名稱詞嵌入的關係。 1.輸入一張圖像，對屬性進行預測 2.根據預測結果，計算屬性詞嵌入的加權平均
相關文章
- 1. [ICCV 17] Learning Discriminative Latent Attributes for Zero-Shot Classification
- 2. Generative model & discriminative model
- 3. DiMP:Learning Discriminative Model Prediction for Tracking
- 4. Stepwise Metric Promotion for Unsupervised Video Person Re-identification
- 5. ICCV 2011
- 6. Dynamic Label Graph Matching for Unsupervised Video Re-Identification
- 7. Learning Discriminative Model Prediction for Tracking --- 論文翻譯
- 8. A Discriminative Feature Learning Approach for Deep Face Recognition
- 9. Learning a Discriminative Feature Network for Semantic Segmentation
- 10. 2017 Discriminative Representation Combinations for Accurate Face Spoofing Detection
- 更多相關文章...
- • Scala for循環 - Scala教程
- • Lua for 循環 - Lua 教程
- • 常用的分佈式事務解決方案
- • 爲了進字節跳動，我精選了29道Java經典算法題，帶詳細講解
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. [ICCV 17] Learning Discriminative Latent Attributes for Zero-Shot Classification
- 2. Generative model & discriminative model
- 3. DiMP:Learning Discriminative Model Prediction for Tracking
- 4. Stepwise Metric Promotion for Unsupervised Video Person Re-identification
- 5. ICCV 2011
- 6. Dynamic Label Graph Matching for Unsupervised Video Re-Identification
- 7. Learning Discriminative Model Prediction for Tracking --- 論文翻譯
- 8. A Discriminative Feature Learning Approach for Deep Face Recognition
- 9. Learning a Discriminative Feature Network for Semantic Segmentation
- 10. 2017 Discriminative Representation Combinations for Accurate Face Spoofing Detection