# Frequency Routing and Identity Conditioning for Thermal-to-Visible Face Translation

arXiv: https://arxiv.org/abs/2610.17859 · HTML: https://mlsanity.com/paper/2610.17859/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 15 references that are in the map.
- First posted on arXiv 13 Oct 2026.

## Builds on (most cited first, 7 of 7)

- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [OminiControl: Minimal and Universal Control for Diffusion Transformer](https://mlsanity.com/paper/2411.15098/) (2024)
- [BBDM: Image-to-image Translation with Brownian Bridge Diffusion Models](https://mlsanity.com/paper/2205.07680/) (2022)
- [Prepended Domain Transformer: Heterogeneous Face Recognition without Bells and Whistles](https://mlsanity.com/paper/2210.06529/) (2022)
- [Thermal to Visible Synthesis of Face Images using Multiple Regions](https://mlsanity.com/paper/1803.07599/) (2018)
- [Polarimetric Thermal to Visible Face Verification via Attribute Preserved Synthesis](https://mlsanity.com/paper/1901.00889/) (2019)
- [MTVDiff: Multimodal Conditional Latent Diffusion for Enhanced Thermal-to-Visible Face Translation](https://mlsanity.com/paper/2607.19886/) (2026)

## Abstract (arXiv)

Thermal-to-visible (T2V) face translation has important applications in nighttime biometric recognition, low-light surveillance, and human interpretation of thermal observations. The task remains challenging because thermal images do not directly record visible appearance, while the cross-spectral cues useful for reconstruction vary across frequency components and semantic roles. Existing spatially conditioned models tend to process these cues through a homogeneous adaptation path and do not explicitly separate local structure from global identity. We propose FreqID-Control, a parameter-efficient framework that organizes thermal guidance according to spectral reliability and semantic role while preserving the pretrained generative prior. It combines frequency-routed low-rank adaptation, which allocates separate transformation subspaces to low-frequency structure and high-frequency detail, with shared cross-spectral identity conditioning, which supplies compact global identity tokens independently of spatial thermal tokens. A controlled analysis on MCXFace and SpeakingFaces provides the empirical basis for this design, revealing stable low-frequency correspondence and edge-localized detail correspondence. Experiments on both datasets demonstrate strong image fidelity and biometric identity preservation, while controlled ablations on MCXFace confirm the complementary contributions of frequency routing and shared identity conditioning.