In recent years, crowd counting has become an important issue in computer vision. In most methods, the density maps are generated by convolving with a Gaussian kernel from the ground-truth dot maps which are marked around the center of human heads. Due to the fixed geometric structures in CNNs and indistinct head-scale information, the head features are obtained incompletely. Deformable Convolution is proposed to exploit the scale-adaptive capabilities for CNN features in the heads. By learning the coordinate offsets of the sampling points, it is tractable to improve the ability to adjust the receptive field. However, the heads are not uniformly covered by the sampling points in the deformable convolution, resulting in loss of head information. To handle the non-uniformed sampling, an improved Normed-Deformable Convolution ( i.e., NDConv) implemented by Normed-Deformable loss ( i.e., NDloss) is proposed in this paper. The offsets of the sampling points which are constrained by NDloss tend to be more even. Then, the features in the heads are obtained more completely, leading to better performance. Especially, the proposed NDConv is a light-weight module which shares similar computation burden with Deformable Convolution. In the extensive experiments, our method outperforms state-of-the-art methods on ShanghaiTech A, ShanghaiTech B, UCF-QNRF, and UCF_CC_50 dataset, achieving 61.4, 7.8, 91.2, and 167.2 MAE, respectively.

中文翻译：

---

用于人群计数的改进的范数可变形卷积

近年来，人群计数已成为计算机视觉中的一个重要问题。在大多数方法中，密度图是通过与标记在人头中心周围的真实点图的高斯核卷积生成的。由于 CNN 中的固定几何结构和不清晰的头部尺度信息，头部特征的获取不完整。提出可变形卷积以利用头部中 CNN 特征的尺度自适应能力。通过学习采样点的坐标偏移量，可以提高调节感受野的能力。然而，在可变形卷积中，头部并没有被采样点均匀覆盖，导致头部信息丢失。为了处理非均匀采样，改进的 Normed-Deformable Convolution ( 即，NDConv) 由 Normed-Deformable loss ( 即，NDloss）在本文中提出。受 NDloss 约束的采样点的偏移量往往更均匀。然后，更完整地获得了头部的特征，从而获得了更好的性能。特别是，所提出的 NDConv 是一个轻量级的模块，它与可变形卷积具有相似的计算负担。在广泛的实验中，我们的方法在 ShanghaiTech A、ShanghaiTech B、UCF-QNRF 和 UCF_CC_50 数据集上优于最先进的方法，分别达到 61.4、7.8、91.2 和 167.2 MAE。