Please enable JavaScript.

Coggle requires JavaScript to display documents.

Single-Image Crowd Counting via Multi-Column Convolutional Neural Network

- - - - In this paper, the viewpoint can be arbitrary => cannot do foreground segmentation
- - - - :bread:output the number of people directly
    - - :bread:image => density map (#people / m^2) => (using integral) #people
      - :pineapple:reasons of recommendation
        
        :one:preserve more information
        
        :question:Don't the interim outputs of the CNN using the first method contain much information, either?
        
        密度图保留更多的信息。与人群的总数相比，密度图给出了在给定图像中人群的空间分布，这样的分布信息在许多应用中是有用的。例如，如果一个小区域的密度比其他区域的密度高得多，它可能表明一些异常发生在那里。
        
        :two:adaptive + heads of different sizes
        
        更适合于有透视效果显着变化的任意输入。所以这些滤波器具有更多的语义，提高了人群计数的准确性
        
        :question:Why?
        
        :explode:the learned filters are more adapted to heads of different sizes
  - - - :question:it is stated that xi is used to represent the pixel position of a head, but how many points does xi represent? Since we know that a head always occupy with many points
      - δ(x−xi)表示图像中人头位置的冲击函
      - N为图像中的人头总数，
  - - - a denstiy map is 4d as well, so it’s okay to use cnn
    - - 输出的密度图 => 标准的密度图
      - loss function
        
        :question:What’s the meaning of double 2
        
        :question:the shape of the Fi?
  - - - 第四层当作最后输出的密度图
      - 分别训练网络的每一层
      - 将每一层参数用了初始化网络并微调
      - :apple:Why
        
        the training dataset is limited
        
        :question:why? Is it because that it is difficult to obtain them? Or there is no ready-made dataset
        
        深度神经网络存在梯度弥散问题
        
        :question:why
  - - - :one:preserve(first several layers) & adapt(last few layers)
      - :two:reduce the computational complexity
- - - - absolute error
      - accuracy of the estimates
    - - mean squared error
      - robustness of the estimates
- - - - :checkered_flag:三列对应不同大小的感应野(大中小),使每个列卷积神经网络的功能对由于透视或不同的图像分辨率造成的人/头大小变化是自适应的（因此，整体网络是强大的）
      - :question:什么是列?
    - - 用一个1*1滤波器的卷积层代替了完全连接的层，因此模型的输入图像可以是任意大小的，避免了失真。网络的直接输出是一个人群密度估计图，从中可以得到的整体计数
      - :question:1*1滤波器怎么代替fc?为什么图像可以是任意大小?Why this can avoid distortion?
    - - 收集了一个新的数据集用于人群计数方法的评价。比现有的数据集包含更复杂的情况，能更好地测试方法性能，1198张图，330,165精确标定的人头。数据集分A和B两个部分，A是从互联网上随机找的图，B是上海的闹市截取图，如图5所示为A、B部分图