# Cross-head Supervision for Crowd Counting with Noisy Annotations

arXiv: https://arxiv.org/abs/2303.09245 · HTML: https://mlsanity.com/paper/2303.09245/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 28 references that are in the map.
- First posted on arXiv 16 Mar 2023.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Early-Learning Regularization Prevents Memorization of Noisy Labels](https://mlsanity.com/paper/2007.00151/) (2020)
- [Unsupervised Label Noise Modeling and Loss Correction](https://mlsanity.com/paper/1904.11238/) (2019)
- [CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes](https://mlsanity.com/paper/1802.10062/) (2018)
- [Composition Loss for Counting, Density Map Estimation and Localization in Dense Crowds](https://mlsanity.com/paper/1808.01050/) (2018)
- [Bayesian Loss for Crowd Count Estimation with Point Supervision](https://mlsanity.com/paper/1908.03684/) (2019)
- [Distribution Matching for Crowd Counting](https://mlsanity.com/paper/2009.13077/) (2020)
- [Boosting Crowd Counting via Multifaceted Attention](https://mlsanity.com/paper/2203.02636/) (2022)
- [Crowd Counting and Density Estimation by Trellis Encoder-Decoder Network](https://mlsanity.com/paper/1903.00853/) (2019)
- [Rethinking Spatial Invariance of Convolutional Networks for Object Counting](https://mlsanity.com/paper/2206.05253/) (2022)
- [Learning Spatial Awareness to Improve Crowd Counting](https://mlsanity.com/paper/1909.07057/) (2019)
- [Weighing Counts: Sequential Crowd Counting by Reinforcement Learning](https://mlsanity.com/paper/2007.08260/) (2020)
- [Twin Contrastive Learning with Noisy Labels](https://mlsanity.com/paper/2303.06930/) (2023)
- [Crowd Counting with Decomposed Uncertainty](https://mlsanity.com/paper/1903.07427/) (2019)
- [PaDNet: Pan-Density Crowd Counting](https://mlsanity.com/paper/1811.02805/) (2018)
- [Uncertainty Estimation and Sample Selection for Crowd Counting](https://mlsanity.com/paper/2009.14411/) (2020)
- [Forget Less, Count Better: A Domain-Incremental Self-Distillation Learning Benchmark for Lifelong Crowd Counting](https://mlsanity.com/paper/2205.03307/) (2022)
- [Multiscale Crowd Counting and Localization By Multitask Point Supervision](https://mlsanity.com/paper/2202.09942/) (2022)

## Built on by (most cited first, 7 of 7)

- [Point, Segment and Count: A Generalized Framework for Object Counting](https://mlsanity.com/paper/2311.12386/) (2023)
- [FGENet: Fine-Grained Extraction Network for Congested Crowd Counting](https://mlsanity.com/paper/2401.01208/) (2024)
- [Diffusion-based Data Augmentation for Object Counting Problems](https://mlsanity.com/paper/2401.13992/) (2024)
- [AdaTreeFormer: Few Shot Domain Adaptation for Tree Counting from a Single High-Resolution Image](https://mlsanity.com/paper/2402.02956/) (2024)
- [Assessing the Visual Enumeration Abilities of Specialized Counting Architectures and Vision-Language Models](https://mlsanity.com/paper/2512.15254/) (2025)
- [GCA-SUNet: A Gated Context-Aware Swin-UNet for Exemplar-Free Counting](https://mlsanity.com/paper/2409.12249/) (2024)
- [EHNet: An Efficient Hybrid Network for Crowd Counting and Localization](https://mlsanity.com/paper/2503.12061/) (2025)

## Abstract (arXiv)

Noisy annotations such as missing annotations and location shifts often exist in crowd counting datasets due to multi-scale head sizes, high occlusion, etc. These noisy annotations severely affect the model training, especially for density map-based methods. To alleviate the negative impact of noisy annotations, we propose a novel crowd counting model with one convolution head and one transformer head, in which these two heads can supervise each other in noisy areas, called Cross-Head Supervision. The resultant model, CHS-Net, can synergize different types of inductive biases for better counting. In addition, we develop a progressive cross-head supervision learning strategy to stabilize the training process and provide more reliable supervision. Extensive experimental results on ShanghaiTech and QNRF datasets demonstrate superior performance over state-of-the-art methods. Code is available at https://github.com/RaccoonDML/CHSNet.