# MTVDiff: Multimodal Conditional Latent Diffusion for Enhanced Thermal-to-Visible Face Translation

arXiv: https://arxiv.org/abs/2607.19886 · HTML: https://mlsanity.com/paper/2607.19886/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- Builds on 25 of its 30 references that are in the map.
- First posted on arXiv 22 Jul 2026.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

## Builds on (most cited first, 25 of 25)

- [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)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://mlsanity.com/paper/2401.10891/) (2024)
- [T2I-Adapter: Learning Adapters to Dig out More Controllable Ability for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.08453/) (2023)
- [Zero-shot Image-to-Image Translation](https://mlsanity.com/paper/2302.03027/) (2023)
- [Exploiting Diffusion Prior for Real-World Image Super-Resolution](https://mlsanity.com/paper/2305.07015/) (2023)
- [Towards Principled Methods for Training Generative Adversarial Networks](https://mlsanity.com/paper/1701.04862/) (2017)
- [Uni-ControlNet: All-in-One Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2305.16322/) (2023)
- [BBDM: Image-to-image Translation with Brownian Bridge Diffusion Models](https://mlsanity.com/paper/2205.07680/) (2022)
- [DiffI2I: Efficient Diffusion Model for Image-to-Image Translation](https://mlsanity.com/paper/2308.13767/) (2023)
- [SpeakingFaces: A Large-Scale Multimodal Dataset of Voice Commands with Visual and Thermal Video Streams](https://mlsanity.com/paper/2012.02961/) (2020)
- [T2V-DDPM: Thermal to Visible Face Translation using Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2209.08814/) (2022)
- [Prepended Domain Transformer: Heterogeneous Face Recognition without Bells and Whistles](https://mlsanity.com/paper/2210.06529/) (2022)
- [DiffV2IR: Visible-to-Infrared Diffusion Model via Vision-Language Understanding](https://mlsanity.com/paper/2503.19012/) (2025)
- [A Diffusion Model Translator for Efficient Image-to-Image Translation](https://mlsanity.com/paper/2502.00307/) (2025)
- [Simultaneous Face Hallucination and Translation for Thermal to Visible Face Verification using Axial-GAN](https://mlsanity.com/paper/2104.06534/) (2021)
- [3D Priors-Guided Diffusion for Blind Face Restoration](https://mlsanity.com/paper/2409.00991/) (2024)
- [SceneDecorator: Towards Scene-Oriented Story Generation with Scene Planning and Scene Consistency](https://mlsanity.com/paper/2510.22994/) (2025)
- [Beyond the Visible: A Survey on Cross-spectral Face Recognition](https://mlsanity.com/paper/2201.04435/) (2022)
- [IdentityStory: Taming Your Identity-Preserving Generator for Human-Centric Story Generation](https://mlsanity.com/paper/2512.23519/) (2025)
- [DiffX: Guide Your Layout to Cross-Modal Generative Modeling](https://mlsanity.com/paper/2407.15488/) (2024)

## Built on by (most cited first, 1 of 1)

- [Frequency Routing and Identity Conditioning for Thermal-to-Visible Face Translation](https://mlsanity.com/paper/2610.17859/) (2026)

## Abstract (arXiv)

Thermal-to-visible face translation presents fundamental challenges including geometric discontinuities, semantic attribute mismatches, and identity degradation. We propose MTVDiff, a novel multimodal latent diffusion framework that synergistically integrates depth and textual information to address these limitations while preserving identity characteristics. The MTVDiff framework presents three core technical contributions: (1) a Dual-Branch Cross-Attention Fusion (DBCAF) module for multi-scale thermal-depth feature extraction and fusion; (2) a Gated Text-to-Visual Feature Alignment mechanism for semantically-guided generation; and (3) Spatial Feature Transformations (SFT) for adaptive multimodal prior integration. Extensive experiments on the MCXFace and SpeakingFaces datasets demonstrate that our multimodal approach significantly outperforms existing GAN-based and diffusion-based approaches across multiple metrics, achieving substantial improvements in both image quality and face verification performance, with FID reductions of up to 48.3% and Rank-1 accuracy improvements of up to 8.9%. Our work provides a robust solution for face recognition systems operating under varying illumination conditions and advances the state-of-the-art in cross-spectral facial image translation through effective multimodal integration.