[Arxiv1706] Few-Example Object Detection with Model Communication 论文笔记

https://arxiv.org/pdf/1706.08249.pdf

Few-Example Object Detection with Model Communication，Xuanyi Dong, Liang Zheng, Fan Ma, Yi Yang, Deyu Meng

 

亮点

• 本文仅仅经过每一个类别3-4个bounding box标注便可实现物体检测，并与其它使用大量training examples的方法性能可比
• 主要方法是：multi-modal learning (多模型同时训练) ＋ self-paced learning (curriculum learning) 

相关工做

这里介绍几个比较容易混淆的概念，以及与他们相关的方法

• 弱监督物体检测：数据集的标签是不可靠的，如（x，y），y对于x的标记是不可靠的。这里的不可靠能够是标记不正确，多种标记，标记不充分，局部标记等。
• 标签是图像级别的类别标签[7][8][9][10][11][18][30][31][32][33][34]
• 半监督物体检测：半监督学习使用大量的未标记数据，以及同时使用标记数据，来进行模式识别工做。
• 一些训练样本只有类别标签，另一些样本有详细的物体框和类别标注[4][5][6]
• 须要大量标注 (e.g., 50% of the full annotations)
• 每一个类别只有几个物体框标注（Few-Example Object Detection with Model Communication)[12][35]
• 和few-shot learning 的区别：是否使用未标注数据学习
• 经过视频挖掘位置标注，此类方法主要针对会移动的物体[2][3][29][1]
• Webly supervised learning for object detection: reduce the annotation cost by leveraging web data

方法

 

 

Basic detector: Faster RCNN & RFCN

Object proposal method: selective search & edge boxes

Annotations: when we randomly annotate approximately four images for each class, an image may contain several objects, and we annotate all the object bounding boxes.

 

参数更新：
更新vj：对上述损失函数进行求导，能够获得vj的解

对同一张图像i同一个模型j，若是有多个样本使得vj＝1，则只选择使Lc最小的那个样本置为1，其余置为0。gamma促使模型之间共享信息，由于vj为1时，阈值变大，图像更容易被选择到。

更新wj：与其它文章方法相同

更新yuj：为更新yuj咱们须要从一组bounding box找到知足如下条件的解，

很难直接找到最优化的解。文中采用的方案是：将全部模型预测出的结果输入nms，并经过阈值只保留分数高的结果，余下的组成yuj。

去除难例：we employ a modified NMS (intersection/max(area1,area2)) to filter out the nested boxes, which usually occurs when there are multiple overlapping objects. If there are too many boxes (≥ 4) for one specific class or too many classes (≥ 4) in the image, this image will be removed. Images in which no reliable pseudo objects are found are filtered out.

实验

Compared with the-state-of-the-art (4.2 images per class is annotated)

• VOC 2007: -1.1mAP, correct localization +0.9% compared with [21]
• VOC 2012: -2.5mAP compared with [21], correct localization +9.8%
• ILSVRC 2013: -2.4mAP compared with [21]
• COCO 2014: +1.3 mAP compared with [22]

[20] V. Kantorov, M. Oquab, M. Cho, and I. Laptev, “Contextlocnet: Context-aware deep network models for weakly supervised localization,” in European Conference on Computer Vision, 2016.
[21] A. Diba, V. Sharma, A. Pazandeh, H. Pirsiavash, and L. Van Gool, “Weakly supervised cascaded convolutional networks,” 2017
[22] Y. Zhu, Y. Zhou, Q. Ye, Q. Qiu, and J. Jiao, “Soft proposal networks for weakly supervised object localization,” in International Conference on Computer Vision, 2017.

Ablation study

• VOC 2007: +4.1 mAP compared with model ensemble
• k number of labeled images per class; w/ image labels: image-level supervision incorporated

　　

 

不足

虽然localization有必定准确率，可是难例图片漏检比较多（也就是说few example classification效果很差）。