# AutoScale: Learning to Scale for Crowd Counting and Localization

arXiv: https://arxiv.org/abs/1912.09632 · HTML: https://mlsanity.com/paper/1912.09632/

## Key facts

- Cited by 35 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Dec 2019.

## Built on by (most cited first, 30 of 35)

- [An End-to-End Transformer Model for Crowd Localization](https://mlsanity.com/paper/2202.13065/) (2022)
- [TransCrowd: weakly-supervised crowd counting with transformers](https://mlsanity.com/paper/2104.09116/) (2021)
- [Rethinking Spatial Invariance of Convolutional Networks for Object Counting](https://mlsanity.com/paper/2206.05253/) (2022)
- [CrowdCLIP: Unsupervised Crowd Counting via Vision-Language Model](https://mlsanity.com/paper/2304.04231/) (2023)
- [Focal Inverse Distance Transform Maps for Crowd Localization](https://mlsanity.com/paper/2102.07925/) (2021)
- [When Counting Meets HMER: Counting-Aware Network for Handwritten Mathematical Expression Recognition](https://mlsanity.com/paper/2207.11463/) (2022)
- [Improving Point-based Crowd Counting and Localization Based on Auxiliary Point Guidance](https://mlsanity.com/paper/2405.10589/) (2024)
- [CrowdDiff: Multi-hypothesis Crowd Density Estimation using Diffusion Models](https://mlsanity.com/paper/2303.12790/) (2023)
- [Counting Crowds in Bad Weather](https://mlsanity.com/paper/2306.01209/) (2023)
- [Boosting Detection in Crowd Analysis via Underutilized Output Features](https://mlsanity.com/paper/2308.16187/) (2023)
- [Crowd Localization from Gaussian Mixture Scoped Knowledge and Scoped Teacher](https://mlsanity.com/paper/2206.05717/) (2022)
- [Congested Crowd Instance Localization with Dilated Convolutional Swin Transformer](https://mlsanity.com/paper/2108.00584/) (2021)
- [YOLC: You Only Look Clusters for Tiny Object Detection in Aerial Images](https://mlsanity.com/paper/2404.06180/) (2024)
- [RGB-D Indiscernible Object Counting in Underwater Scenes](https://mlsanity.com/paper/2304.11677/) (2023)
- [CrowdFormer: Weakly-supervised Crowd counting with Improved Generalizability](https://mlsanity.com/paper/2203.03768/) (2022)
- [Consistency-Aware Anchor Pyramid Network for Crowd Localization](https://mlsanity.com/paper/2212.04067/) (2022)
- [Focus for Free in Density-Based Counting](https://mlsanity.com/paper/2306.05129/) (2023)
- [Learning Discriminative Features for Crowd Counting](https://mlsanity.com/paper/2311.04509/) (2023)
- [VisDrone-CC2020: The Vision Meets Drone Crowd Counting Challenge Results](https://mlsanity.com/paper/2107.08766/) (2021)
- [Progressive Multi-resolution Loss for Crowd Counting](https://mlsanity.com/paper/2212.04127/) (2022)
- [Mask Focal Loss: A unifying framework for dense crowd counting with canonical object detection networks](https://mlsanity.com/paper/2212.11542/) (2022)
- [Counting Like Human: Anthropoid Crowd Counting on Modeling the Similarity of Objects](https://mlsanity.com/paper/2212.02248/) (2022)
- [Super-Resolution Information Enhancement For Crowd Counting](https://mlsanity.com/paper/2303.06925/) (2023)
- [Shifted Autoencoders for Point Annotation Restoration in Object Counting](https://mlsanity.com/paper/2312.07190/) (2023)
- [Tree Counting by Bridging 3D Point Clouds with Imagery](https://mlsanity.com/paper/2403.01932/) (2024)
- [Skim then Focus: Integrating Contextual and Fine-grained Views for Repetitive Action Counting](https://mlsanity.com/paper/2406.08814/) (2024)
- [Exploring Scale Shift in Crowd Localization under the Context of Domain Generalization](https://mlsanity.com/paper/2510.19330/) (2025)
- [Indirect-Instant Attention Optimization for Crowd Counting in Dense Scenes](https://mlsanity.com/paper/2206.05648/) (2022)
- [Scale-Aware Crowd Counting Using a Joint Likelihood Density Map and Synthetic Fusion Pyramid Network](https://mlsanity.com/paper/2211.06835/) (2022)
- [Scale-Aware Crowd Count Network with Annotation Error Correction](https://mlsanity.com/paper/2312.16771/) (2023)

## Abstract (arXiv)

Recent works on crowd counting mainly leverage CNNs to count by regressing density maps, and have achieved great progress. In the density map, each person is represented by a Gaussian blob, and the final count is obtained from the integration of the whole map. However, it is difficult to accurately predict the density map on dense regions. A major issue is that the density map on dense regions usually accumulates density values from a number of nearby Gaussian blobs, yielding different large density values on a small set of pixels. This makes the density map present variant patterns with significant pattern shifts and brings a long-tailed distribution of pixel-wise density values. We propose a simple and effective Learning to Scale (L2S) module, which automatically scales dense regions into reasonable closeness levels (reflecting image-plane distance between neighboring people). L2S directly normalizes the closeness in different patches such that it dynamically separates the overlapped blobs, decomposes the accumulated values in the ground-truth density map, and thus alleviates the pattern shifts and long-tailed distribution of density values. This helps the model to better learn the density map. We also explore the effectiveness of L2S in localizing people by finding the local minima of the quantized distance (w.r.t. person location map). To the best of our knowledge, such a localization method is also novel in localization-based crowd counting. We further introduce a customized dynamic cross-entropy loss, significantly improving the localization-based model optimization. Extensive experiments demonstrate that the proposed framework termed AutoScale improves upon some state-of-the-art methods in both regression and localization benchmarks on three crowded datasets and achieves very competitive performance on two sparse datasets.