Deep patch learning for weakly supervised object classification and discovery
作者：
Highlights：
• We propose to integrate different patch-based object classification stages into a weakly supervised deep CNN framework.
• We integrate the two MIL constraints into the loss of our deep CNN framework for object discovery.
• We embed object classification and discovery into a multi-task CNN, and demonstrate they are complementary.
• Our method DPL learns patch features end-to-end, and is more effective and efficient than previous patch-based CNNs.
• DPL obtains state-of-the-art results on classification and competitive results on discovery, with fast testing speed.
摘要
•We propose to integrate different patch-based object classification stages into a weakly supervised deep CNN framework.•We integrate the two MIL constraints into the loss of our deep CNN framework for object discovery.•We embed object classification and discovery into a multi-task CNN, and demonstrate they are complementary.•Our method DPL learns patch features end-to-end, and is more effective and efficient than previous patch-based CNNs.•DPL obtains state-of-the-art results on classification and competitive results on discovery, with fast testing speed.
论文关键词：Patch feature learning,Multiple instance learning,Weakly supervised learning,Convolutional neural network,End-to-end,Object classification,Object discovery
论文评审过程：Received 31 May 2016, Revised 12 April 2017, Accepted 2 May 2017, Available online 3 May 2017, Version of Record 12 July 2017.
论文官网地址：https://doi.org/10.1016/j.patcog.2017.05.001
