Distilling Object Detectors with Fine-grained Feature Imitation
Motivation 檢測起更focus在物體出現的區域 Detectors care more about local near object regions. 物體出現的周圍特徵變化其實包含了更多重要信息，這是student網絡需要向teacher網絡學習的 註解： 與分類不同，蒸餾方法在檢測中如果進行全特徵模仿的話對子網絡的提升很有限(這裏存疑，文章沒有明確指出全特徵模仿了哪些特徵層)。 這
相關文章
- 1. 【Distill 系列：二】CVPR 2019 Distilling Object Detectors with Fine-grained Feature Imitation
- 2. OHEM：Training Region-based Object Detectors with Online Hard Example Mining
- 3. Training Region-based Object Detectors with Online Hard Example Mining
- 4. one-stage object detectors(1)
- 5. 論文閱讀：DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
- 6. Reading Note: DSOD: Learning Deeply Supervised Object Detectors from Scratch
- 7. diagnosing error in object detectors 淺析
- 8. 【論文筆記】：DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
- 9. CVPR 2017:Large Margin Object Tracking with Circulant Feature Maps
- 10. DSOD: Learning Deeply Supervised Object Detectors from Scratch
- 更多相關文章...
- • XSL-FO instream-foreign-object 對象 - XSL-FO 教程
- • XSLT
- • 爲了進字節跳動，我精選了29道Java經典算法題，帶詳細講解
- • JDK13 GA發佈：5大特性解讀
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. 【Distill 系列：二】CVPR 2019 Distilling Object Detectors with Fine-grained Feature Imitation
- 2. OHEM：Training Region-based Object Detectors with Online Hard Example Mining
- 3. Training Region-based Object Detectors with Online Hard Example Mining
- 4. one-stage object detectors(1)
- 5. 論文閱讀：DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
- 6. Reading Note: DSOD: Learning Deeply Supervised Object Detectors from Scratch
- 7. diagnosing error in object detectors 淺析
- 8. 【論文筆記】：DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
- 9. CVPR 2017:Large Margin Object Tracking with Circulant Feature Maps
- 10. DSOD: Learning Deeply Supervised Object Detectors from Scratch