# Thermal-Only Crowd Counting with Deployment-Time Privacy Protection

arXiv: https://arxiv.org/abs/2605.17042 · HTML: https://mlsanity.com/paper/2605.17042/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 32 of its 58 references that are in the map.
- First posted on arXiv 16 May 2026.

## Builds on (most cited first, 30 of 32)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions](https://mlsanity.com/paper/2102.12122/) (2021)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [Extracting Training Data from Diffusion Models](https://mlsanity.com/paper/2301.13188/) (2023)
- [Label-Efficient Semantic Segmentation with Diffusion Models](https://mlsanity.com/paper/2112.03126/) (2021)
- [Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion](https://mlsanity.com/paper/2310.02279/) (2023)
- [Improved Techniques for Training Consistency Models](https://mlsanity.com/paper/2310.14189/) (2023)
- [CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes](https://mlsanity.com/paper/1802.10062/) (2018)
- [BBDM: Image-to-image Translation with Brownian Bridge Diffusion Models](https://mlsanity.com/paper/2205.07680/) (2022)
- [LCM-LoRA: A Universal Stable-Diffusion Acceleration Module](https://mlsanity.com/paper/2311.05556/) (2023)
- [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)
- [UNIT-DDPM: UNpaired Image Translation with Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2104.05358/) (2021)
- [Semantic Image Synthesis via Diffusion Models](https://mlsanity.com/paper/2207.00050/) (2022)
- [Boosting Crowd Counting via Multifaceted Attention](https://mlsanity.com/paper/2203.02636/) (2022)
- [Cross-Modal Collaborative Representation Learning and a Large-Scale RGBT Benchmark for Crowd Counting](https://mlsanity.com/paper/2012.04529/) (2020)
- [PCC Net: Perspective Crowd Counting via Spatial Convolutional Network](https://mlsanity.com/paper/1905.10085/) (2019)
- [Direct Measure Matching for Crowd Counting](https://mlsanity.com/paper/2107.01558/) (2021)
- [TAFNet: A Three-Stream Adaptive Fusion Network for RGB-T Crowd Counting](https://mlsanity.com/paper/2202.08517/) (2022)
- [Counting Crowds in Bad Weather](https://mlsanity.com/paper/2306.01209/) (2023)
- [Spatio-channel Attention Blocks for Cross-modal Crowd Counting](https://mlsanity.com/paper/2210.10392/) (2022)
- [CLIP-EBC: CLIP Can Count Accurately through Enhanced Blockwise Classification](https://mlsanity.com/paper/2403.09281/) (2024)
- [Multi-modal Crowd Counting via a Broker Modality](https://mlsanity.com/paper/2407.07518/) (2024)
- [Focus for Free in Density-Based Counting](https://mlsanity.com/paper/2306.05129/) (2023)
- [RGB-T Multi-Modal Crowd Counting Based on Transformer](https://mlsanity.com/paper/2301.03033/) (2023)

## Abstract (arXiv)

While RGB-Thermal crowd counting has shown promise, the paradigm faces critical limitations: RGB data raises privacy concerns in public surveillance, and multi-modal misalignment degrades fusion performance. We propose the first thermal-only framework specifically designed for privacy-conscious crowd counting, eliminating RGB dependency at inference time and substantially reducing the privacy exposure associated with continuous RGB capture in public surveillance deployments. To mitigate thermal ambiguity, we leverage depth-to-RGB diffusion models as a cross-modal bridge, extracting discriminative features that enhance thermal representations. Critically, we demonstrate that single-step LCM denoising yields features most faithful to the structural content of the depth conditioning signal, while multi-step approaches progressively decouple features from the conditioning input and accumulate errors that degrade counting accuracy. Experiments on RGBT-CC and DroneRGBT datasets show our method achieves competitive performance against state-of-the-art RGB-T fusion methods, while requiring only thermal input during inference, eliminating the need for continuous RGB capture that constitutes the primary privacy concern in real-world surveillance deployment. The code will be made publicly available.