mlsanityData to 14 Oct 2026

SIRR-LMM: Single-image Reflection Removal via Large Multimodal Model

arXiv 2601.07209 ↗First posted 12 Jan 2026cs.CV cs.AI cs.GRcited by 0

On the map: Multimodal LLMs / vision-language models → MLLM training and visual instruction tuning

Abstract (arXiv)

Glass surfaces create complex interactions of reflected and transmitted light, making single-image reflection removal (SIRR) challenging. Existing datasets suffer from limited physical realism in synthetic data or insufficient scale in real captures. We introduce a synthetic dataset generation framework that path-traces 3D glass models over real background imagery to create physically accurate reflection scenarios with varied glass properties, camera settings, and post-processing effects. To leverage the capabilities of Large Multimodal Model (LMM), we concatenate the image layers into a single composite input, apply joint captioning, and fine-tune the model using task-specific LoRA rather than full-parameter training. This enables our approach to achieve improved reflection removal and separation performance compared to state-of-the-art methods.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 30 of its 63 references that are in the map. First posted on arXiv 12 Jan 2026.

Builds on

30 of 63 references.

References · 30

  1. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  2. 2022Scalable Diffusion Models with Transformerscited by 6,054
  3. 2025FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Spacecited by 1,006
  4. 2024In-Context LoRA for Diffusion Transformerscited by 158
  5. 2018Single Image Reflection Separation with Perceptual Lossescited by 98
  6. 2020Polarized Reflection Removal with Perfect Alignment in the Wildcited by 69
  7. 2017A Generic Deep Architecture for Single Image Reflection Removal and Image Smoothingcited by 60
  8. 2019Single Image Reflection Removal through Cascaded Refinementcited by 58
  9. 2019Single Image Reflection Removal Exploiting Misaligned Training Data and Network Enhancementscited by 54
  10. 2020Location-aware Single Image Reflection Removalcited by 43
  11. 2020Learning to See Through Obstructionscited by 38
  12. 2023Single Image Reflection Separation via Component Synergycited by 38
18 more
  1. 2021Robust Reflection Removal with Reflection-free Flash-only Cuescited by 34
  2. 2023Revisiting Single Image Reflection Removal In the Wildcited by 33
  3. 2021Trash or Treasure? An Interactive Dual-Stream Strategy for Single Image Reflection Separationcited by 32
  4. 2018CRRN: Multi-Scale Guided Concurrent Reflection Removal Networkcited by 24
  5. 2020Two-Stage Single Image Reflection Removal with Reflection-Aware Guidancecited by 22
  6. 2024Reversible Decoupling Network for Single Image Reflection Removalcited by 22
  7. 2017Separating Reflection and Transmission Images in the Wildcited by 19
  8. 2024Language-guided Image Reflection Separationcited by 16
  9. 2022Robust Reflection Removal with Flash-only Cues in the Wildcited by 14
  10. 2021A Categorized Reflection Removal Dataset with Diverse Real-world Scenescited by 12
  11. 2025Dereflection Any Image with Diffusion Priors and Diversified Datacited by 12
  12. 2024Removing Reflections from RAW Photoscited by 10
  13. 2024Flash-Split: 2D Reflection Removal with Flash Cues and Latent Diffusion Separationcited by 10
  14. 2025Survey on Single-Image Reflection Removal using Deep Learning Techniquescited by 6
  15. 2019Semantic Guided Single Image Reflection Removalcited by 4
  16. 2024FIRM: Flexible Interactive Reflection reMovalcited by 4
  17. 2025ePBR: Extended PBR Materials in Image Synthesiscited by 3
  18. 2025F2T2-HiT: A U-Shaped FFT Transformer and Hierarchical Transformer for Reflection Removalcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Reflection Removal through Efficient Adaptation of Diffusion Transformers
  2. 2024Single-image reflection removal via self-supervised diffusion models
  3. 2026PRISM: Latent Composition Consistency for Single-Image Reflection Removal
  4. 2026FUMO: Prior-Modulated Diffusion for Single Image Reflection Removal
  5. 2026From Synthesis to Removal: Physics-Grounded Reflection Simulation and Diffusion-Based Video Dereflection
  6. 2026ReLo-IRR: Reflection-Guided LoRA Framework for Image Reflection Removal
  7. 2026LowAux-RDNet: Low-Pass Residual Supervision with Scene-Balanced Real-World Training for Single-Image Reflection Removal
  8. 2024PromptRR: Diffusion Models as Prompt Generators for Single Image Reflection Removal