mlsanityData to 7 Oct 2026

MagicQuillV2: Precise and Interactive Image Editing with Layered Visual Cues

arXiv 2512.03046 ↗First posted 2 Dec 2025cs.CVcited by 4

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

We propose MagicQuill V2, a novel system that introduces a layered composition paradigm to generative image editing, bridging the gap between the semantic power of diffusion models and the granular control of traditional graphics software. While diffusion transformers excel at holistic generation, their use of singular, monolithic prompts fails to disentangle distinct user intentions for content, position, and appearance. To overcome this, our method deconstructs creative intent into a stack of controllable visual cues: a content layer for what to create, a spatial layer for where to place it, a structural layer for how it is shaped, and a color layer for its palette. Our technical contributions include a specialized data generation pipeline for context-aware content integration, a unified control module to process all visual cues, and a fine-tuned spatial branch for precise local editing, including object removal. Extensive experiments validate that this layered approach effectively resolves the user intention gap, granting creators direct, intuitive control over the generative process.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 65 of its 82 references that are in the map. First posted on arXiv 2 Dec 2025.

Builds on

65 of 82 references.

Closest ancestors · 13

Same direction, newest first

  1. 2025FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space17 Jun
  2. 2025OminiControl2: Efficient Conditioning for Diffusion Transformers11 Mar
  3. 2024LayerFusion: Harmonized Multi-Layer Text-to-Image Generation with Generative Priors5 Dec
  4. 2024Diffusion Self-Distillation for Zero-Shot Customized Image Generation27 Nov
  5. 2024Generating Compositional Scenes via Text-to-image RGBA Instance Generation16 Nov
  6. 2024ControlAR: Controllable Image Generation with Autoregressive Models3 Oct
  7. 2024FreeCompose: Generic Zero-Shot Image Composition with Diffusion Prior6 Jul
  8. 2024LayerDiff: Exploring Text-guided Multi-layered Composable Image Synthesis via Layer-Collaborative Diffusion Model18 Mar
  9. 2023TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition24 Jul
  10. 2023Text2Layer: Layered Image Generation using Latent Diffusion Model19 Jul
  11. 2023T2I-Adapter: Learning Adapters to Dig out More Controllable Ability for Text-to-Image Diffusion Models16 Feb
  12. 2023Adding Conditional Control to Text-to-Image Diffusion Models10 Feb
1 more
  1. 2022Paint by Example: Exemplar-based Image Editing with Diffusion Models23 Nov

Other references · 52

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2025Qwen3 Technical Reportcited by 8,195
  4. 2023Segment Anythingcited by 8,158
  5. 2022Scalable Diffusion Models with Transformerscited by 5,931
  6. 2025Qwen2.5-VL Technical Reportcited by 5,418
  7. 2025Qwen-Image Technical Reportcited by 973
  8. 2024Grounded SAM: Assembling Open-World Models for Diverse Visual Taskscited by 914
  9. 2023Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inferencecited by 826
  10. 2015Holistically-Nested Edge Detectioncited by 590
  11. 2025VACE: All-in-One Video Creation and Editingcited by 467
  12. 2025Step1X-Edit: A Practical Framework for General Image Editingcited by 417
38 more
  1. 2023AnyDoor: Zero-shot Object-level Image Customizationcited by 411
  2. 2024OminiControl: Minimal and Universal Control for Diffusion Transformercited by 334
  3. 2024AnyEdit: Mastering Unified High-Quality Image Editing for Any Ideacited by 235
  4. 2025Seedream 4.0: Toward Next-generation Multimodal Image Generationcited by 233
  5. 2023LCM-LoRA: A Universal Stable-Diffusion Acceleration Modulecited by 211
  6. 2023Guiding Instruction-based Image Editing via Multimodal Large Language Modelscited by 182
  7. 2021Pixel Difference Networks for Efficient Edge Detectioncited by 167
  8. 2024SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Modelscited by 161
  9. 2019DoveNet: Deep Image Harmonization via Domain Verificationcited by 134
  10. 2017Deep Image Harmonizationcited by 120
  11. 2025EasyControl: Adding Efficient and Flexible Control for Diffusion Transformercited by 118
  12. 2024UniReal: Universal Image Generation and Editing via Learning Real-world Dynamicscited by 100
  13. 2025ACE++: Instruction-Based Image Creation and Editing via Context-Aware Content Fillingcited by 95
  14. 2022Learning to generate line drawings that convey geometry and semanticscited by 82
  15. 2021Region-aware Adaptive Instance Normalization for Image Harmonizationcited by 71
  16. 2024IMPRINT: Generative Object Compositing by Learning Identity-Preserving Representationcited by 64
  17. 2024Zero-shot Image Editing with Reference Imitationcited by 64
  18. 2021Making Images Real Again: A Comprehensive Survey on Deep Image Compositioncited by 63
  19. 2025Insert Anything: Image Insertion via In-Context Editing in DiTcited by 61
  20. 2021SSH: A Self-Supervised Framework for Image Harmonizationcited by 60
  21. 2025Pico-Banana-400K: A Large-Scale Dataset for Text-Guided Image Editingcited by 58
  22. 2020Foreground-aware Semantic Representations for Image Harmonizationcited by 53
  23. 2015Learning a Discriminative Model for the Perception of Realism in Composite Imagescited by 52
  24. 2020BargainNet: Background-Guided Domain Translation for Image Harmonizationcited by 52
  25. 2018Semi-parametric Image Synthesiscited by 50
  26. 2025EditAR: Unified Conditional Generation with Autoregressive Modelscited by 40
  27. 2024MULAN: A Multi Layer Annotated Dataset for Controllable Text-to-Image Generationcited by 38
  28. 2025OmniEraser: Remove Objects and Their Effects in Images with Paired Video-Frame Datacited by 35
  29. 2024Generative Image Layer Decomposition with Visual Effectscited by 33
  30. 2024MagicQuill: An Intelligent Interactive Image Editing Systemcited by 30
  31. 2023Tiny and Efficient Model for the Edge Detection Generalizationcited by 28
  32. 2025SmartEraser: Remove Anything from Images using Masked-Region Guidancecited by 24
  33. 2024Alfie: Democratising RGBA Image Generation With No cited by 20
  34. 2025RelaCtrl: Relevance-Guided Efficient Control for Diffusion Transformerscited by 18
  35. 2025LayeringDiff: Layered Image Synthesis via Generation, then Disassembly with Generative Knowledgecited by 15
  36. 2025PSDiffusion: Harmonized Multi-Layer Image Generation via Layout and Appearance Alignmentcited by 11
  37. 2025IC-Custom: Diverse Image Customization via In-Context Learningcited by 10
  38. 2025AI-Instruments: Embodying Prompts as Instruments to Abstract & Reflect Graphical Interface Commands as General-Purpose Toolscited by 8

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026Banana100: Breaking NR-IQA Metrics by 100 Iterative Image Replications with Nano Banana Procited by 1
  2. 2026Does Synthetic Layered Design Data Benefit Layered Design Decomposition?cited by 1
  3. 2026ExpressEdit: Fast Editing of Stylized Facial Expressions with Diffusion Models in Photoshopcited by 0
  4. 2026LimeCross: Context-Conditioned Layered Image Editing with Structural Consistencycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Editing on the Generative Manifold: A Theoretical and Empirical Study of General Diffusion-Based Image Editing Trade-offs
  2. 2026MRT: Masked Region Transformer for Layered Image Generation and Editing at Scale
  3. 2024Streamlining Image Editing with Layered Diffusion Brushes
  4. 2024Instruction-Guided Editing Controls for Images and Multimedia: A Survey in LLM era
  5. 2025ReMix: Towards a Unified View of Consistent Character Generation and Editing
  6. 2026How Well Do Models Follow Visual Instructions? VIBE: A Systematic Benchmark for Visual Instruction-Driven Image Editing
  7. 2025BlenderFusion: 3D-Grounded Visual Editing and Generative Compositing
  8. 2025LayerCraft: Enhancing Text-to-Image Generation with CoT Reasoning and Layered Object Integration