# UnReflectAnything: RGB-Only Highlight Removal by Rendering Synthetic Specular Supervision

arXiv: https://arxiv.org/abs/2512.09583 · HTML: https://mlsanity.com/paper/2512.09583/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 38 references that are in the map.
- First posted on arXiv 10 Dec 2025.

## Builds on (most cited first, 17 of 17)

- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [LightGlue: Local Feature Matching at Light Speed](https://mlsanity.com/paper/2306.13643/) (2023)
- [EndoNet: A Deep Architecture for Recognition Tasks on Laparoscopic Videos](https://mlsanity.com/paper/1602.03012/) (2016)
- [DISK: Learning local features with policy gradient](https://mlsanity.com/paper/2006.13566/) (2020)
- [MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Details](https://mlsanity.com/paper/2507.02546/) (2025)
- [MAGSAC++, a fast, reliable and accurate robust estimator](https://mlsanity.com/paper/1912.05909/) (2019)
- [StableNormal: Reducing Diffusion Variance for Stable and Sharp Normal](https://mlsanity.com/paper/2406.16864/) (2024)
- [Stereo Correspondence and Reconstruction of Endoscopic Data Challenge](https://mlsanity.com/paper/2101.01133/) (2021)
- [HouseCat6D -- A Large-Scale Multi-Modal Category Level 6D Object Perception Dataset with Household Objects in Realistic Scenarios](https://mlsanity.com/paper/2212.10428/) (2022)
- [Learning How To Robustly Estimate Camera Pose in Endoscopic Videos](https://mlsanity.com/paper/2304.08023/) (2023)
- [Dual-Hybrid Attention Network for Specular Highlight Removal](https://mlsanity.com/paper/2407.12255/) (2024)
- [PolarFree: Polarization-based Reflection-free Imaging](https://mlsanity.com/paper/2503.18055/) (2025)
- [CroMo: Cross-Modal Learning for Monocular Depth Estimation](https://mlsanity.com/paper/2203.12485/) (2022)
- [SCRREAM : SCan, Register, REnder And Map:A Framework for Annotating Accurate and Dense 3D Indoor Scenes with a Benchmark](https://mlsanity.com/paper/2410.22715/) (2024)
- [RIDE: Self-Supervised Learning of Rotation-Equivariant Keypoint Detection and Invariant Description for Endoscopy](https://mlsanity.com/paper/2309.09563/) (2023)
- [LiteTracker: Leveraging Temporal Causality for Accurate Low-latency Tissue Tracking](https://mlsanity.com/paper/2504.09904/) (2025)
- [PolarAnything: Diffusion-based Polarimetric Image Synthesis](https://mlsanity.com/paper/2507.17268/) (2025)

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

- [Ocular Verification for Virtual Reality](https://mlsanity.com/paper/2607.20790/) (2026)

## Abstract (arXiv)

Specular highlights distort appearance, obscure texture, and hinder geometric reasoning in both natural and surgical imagery. We present UnReflectAnything, an RGB-only framework that removes highlights from a single image by predicting a highlight map together with a reflection-free diffuse reconstruction. The model uses a frozen vision transformer encoder to extract multi-scale features, a lightweight head to localize specular regions, and a token-level inpainting module that restores corrupted feature patches before producing the final diffuse image. To overcome the lack of paired supervision, we introduce a Virtual Highlight Synthesis pipeline that renders physically plausible specularities using monocular geometry, Fresnel-aware shading, and randomized lighting which enables training on arbitrary RGB images with correct geometric structure. UnReflectAnything generalizes across natural and surgical domains where non-Lambertian surfaces and non-uniform lighting create severe highlights and it achieves competitive performance with state-of-the-art results on several benchmarks. Project Page: https://alberto-rota.github.io/UnReflectAnything/