# TAFNet: A Three-Stream Adaptive Fusion Network for RGB-T Crowd Counting

arXiv: https://arxiv.org/abs/2202.08517 · HTML: https://mlsanity.com/paper/2202.08517/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 24 references that are in the map.
- First posted on arXiv 17 Feb 2022.

## Builds on (most cited first, 12 of 12)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes](https://mlsanity.com/paper/1802.10062/) (2018)
- [Bayesian Loss for Crowd Count Estimation with Point Supervision](https://mlsanity.com/paper/1908.03684/) (2019)
- [CrowdNet: A Deep Convolutional Network for Dense Crowd Counting](https://mlsanity.com/paper/1608.06197/) (2016)
- [Cross-Modal Collaborative Representation Learning and a Large-Scale RGBT Benchmark for Crowd Counting](https://mlsanity.com/paper/2012.04529/) (2020)
- [A Unified RGB-T Saliency Detection Benchmark: Dataset, Baselines, Analysis and A Novel Approach](https://mlsanity.com/paper/1701.02829/) (2017)
- [CNN-based Density Estimation and Crowd Counting: A Survey](https://mlsanity.com/paper/2003.12783/) (2020)
- [A Self-Training Approach for Point-Supervised Object Detection and Counting in Crowds](https://mlsanity.com/paper/2007.12831/) (2020)

## Built on by (most cited first, 15 of 15)

- [Infrared and Visible Image Fusion: From Data Compatibility to Task Adaption](https://mlsanity.com/paper/2501.10761/) (2025)
- [Explicit Attention-Enhanced Fusion for RGB-Thermal Perception Tasks](https://mlsanity.com/paper/2303.15710/) (2023)
- [Revisiting Crowd Counting: State-of-the-art, Trends, and Future Perspectives](https://mlsanity.com/paper/2209.07271/) (2022)
- [Multi-modal Crowd Counting via a Broker Modality](https://mlsanity.com/paper/2407.07518/) (2024)
- [RGB-T Multi-Modal Crowd Counting Based on Transformer](https://mlsanity.com/paper/2301.03033/) (2023)
- [CLIP: Train Faster with Less Data](https://mlsanity.com/paper/2212.01452/) (2022)
- [DroneNet: Crowd Density Estimation using Self-ONNs for Drones](https://mlsanity.com/paper/2211.07137/) (2022)
- [ComPtr: Towards Diverse Bi-source Dense Prediction Tasks via A Simple yet General Complementary Transformer](https://mlsanity.com/paper/2307.12349/) (2023)
- [Multi-modal Crowd Counting via Modal Emulation](https://mlsanity.com/paper/2407.19491/) (2024)
- [Why Existing Multimodal Crowd Counting Datasets Can Lead to Unfulfilled Expectations in Real-World Applications](https://mlsanity.com/paper/2304.06401/) (2023)
- [Depth-Assisted Network for Indiscernible Marine Object Counting with Adaptive Motion-Differentiated Feature Encoding](https://mlsanity.com/paper/2503.08152/) (2025)
- [MAFNet: A Multi-Attention Fusion Network for RGB-T Crowd Counting](https://mlsanity.com/paper/2208.06761/) (2022)
- [Crowd Density Estimation using Imperfect Labels](https://mlsanity.com/paper/2212.01450/) (2022)
- [Thermal-Only Crowd Counting with Deployment-Time Privacy Protection](https://mlsanity.com/paper/2605.17042/) (2026)
- [Multi-Modal Hyper-Graph Fusion for Low-Light Crowd Counting](https://mlsanity.com/paper/2606.18566/) (2026)

## Abstract (arXiv)

In this paper, we propose a three-stream adaptive fusion network named TAFNet, which uses paired RGB and thermal images for crowd counting. Specifically, TAFNet is divided into one main stream and two auxiliary streams. We combine a pair of RGB and thermal images to constitute the input of main stream. Two auxiliary streams respectively exploit RGB image and thermal image to extract modality-specific features. Besides, we propose an Information Improvement Module (IIM) to fuse the modality-specific features into the main stream adaptively. Experiment results on RGBT-CC dataset show that our method achieves more than 20% improvement on mean average error and root mean squared error compared with state-of-the-art method. The source code will be publicly available at https://github.com/TANGHAIHAN/TAFNet.