亮點

• 一個好名字給了讓讀者開始閱讀的理由
• global max pooling over sliding window的定位方法值得借鑒

方法

本文的目標是：設計一個弱監督分類網絡，註意本文的目標主要是提升分類。因為是2015年的文章，方法比較簡單原始。

Following three modifications to a classification network.

• Treat the fully connected layers as convolutions, which allows us to deal with nearly arbitrary-sized images as input.
• The aim is to apply the network to bigger images in a sliding window manner thus extending its output to n×m× K, where n and m denote the number of sliding window positions in the x- and y- direction in the image, respectively.
• 3xhxw —> convs —> kxmxn (k: number of classes)
• Explicitly search for the highest scoring object position in the image by adding a single global max-pooling layer at the output.
• kxmxn —> kx1x1
• The max-pooling operation hypothesizes the location of the object in the image at the position with the maximum score
• Use a cost function that can explicitly model multiple objects present in the image.

因為圖中可能有很多物體，所以多類的分類loss不適用。作者把這個任務視為多個二分類問題，loss function和分類的分數如下

training

muti-scale test

實驗

classification

• mAP on VOC 2012 test: ＋3.1% compared with [56]
• mAP on VOC 2012 test: ＋7.6% compared with kx1x1 output and single scale training
• mAP on VOC: ＋2.6% compared with RCNN
• mAP on COCO 62.8%

Localisation

• Metric: if the maximal response across scales falls within the ground truth bounding box of an object of the same class within 18 pixels tolerance, we label the predicted location as correct. If not, then we count the response as a false positive (it hit the background), and we also increment the false negative count (no object was found).
• metric on VOC 2012 val: -0.3% compared with RCNN
• mAP on COCO 41.2%

缺點

• 定位評測的metric不具有權威性
• max pooling改為average pooling會不會對於多個instance的情況更好一些

[CVPR2015] Is object localization for free? – Weakly-supervised learning with convolutional neural networks論文筆記