mlsanityData to 7 Oct 2026

Learning a Discriminative Model for the Perception of Realism in Composite Images

arXiv 1510.00477 ↗First posted 2 Oct 2015cs.CVcited by 52

Not on the map. Why?

Abstract (arXiv)

What makes an image appear realistic? In this work, we are answering this question from a data-driven perspective by learning the perception of visual realism directly from large amounts of data. In particular, we train a Convolutional Neural Network (CNN) model that distinguishes natural photographs from automatically generated composite images. The model learns to predict visual realism of a scene in terms of color, lighting and texture compatibility, without any human annotations pertaining to it. Our model outperforms previous works that rely on hand-crafted heuristics, for the task of classifying realistic vs. unrealistic photos. Furthermore, we apply our learned model to compute optimal parameters of a compositing method, to maximize the visual realism score predicted by our CNN model. We demonstrate its advantage against existing methods via a human perception study.

Key facts. Cited by 52 papers in the map (data through 7 Oct 2026). First posted on arXiv 2 Oct 2015.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 12; 2022: 14; 2023: 15; 2024: 6; 2025: 3; 2026: 2.

52 papers.

Sort:

By area:

  1. 2024Transparent Image Layer Diffusion using Latent Transparencycited by 111
  2. 2021Region-aware Adaptive Instance Normalization for Image Harmonizationcited by 71
  3. 2021Making Images Real Again: A Comprehensive Survey on Deep Image Compositioncited by 63
  4. 2022Harmonizer: Learning to Perform White-Box Image and Video Harmonizationcited by 61
  5. 2021SSH: A Self-Supervised Framework for Image Harmonizationcited by 60
  6. 2023Relightful Harmonization: Lighting-aware Portrait Background Replacementcited by 56
  7. 2021CAMPARI: Camera-Aware Decomposed Generative Neural Radiance Fieldscited by 55
  8. 2021High-Resolution Image Harmonization via Collaborative Dual Transformationscited by 55
  9. 2021TransFill: Reference-guided Image Inpainting by Merging Multiple Color and Spatial Transformationscited by 48
  10. 2021Shadow Generation for Composite Image in Real-world Scenescited by 35
  11. 2023Semi-supervised Parametric Real-world Image Harmonizationcited by 26
  12. 2021Spatial-Separated Curve Rendering Network for Efficient and High-Resolution Image Harmonizationcited by 25
38 more
  1. 2022SCS-Co: Self-Consistent Style Contrastive Learning for Image Harmonizationcited by 24
  2. 2022Learning Object Placement via Dual-path Graph Completioncited by 20
  3. 2023Towards Enhanced Image Inpainting: Mitigating Unwanted Object Insertion and Preserving Color Consistencycited by 19
  4. 2023NCHO: Unsupervised Learning for Neural 3D Composition of Humans and Objectscited by 17
  5. 2024FreeCompose: Generic Zero-Shot Image Composition with Diffusion Priorcited by 16
  6. 2023Realistic Saliency Guided Image Enhancementcited by 14
  7. 20233D-aware Blending with Generative NeRFscited by 13
  8. 2023LEMaRT: Label-Efficient Masked Region Transform for Image Harmonizationcited by 12
  9. 2023Painterly Image Harmonization using Diffusion Modelcited by 12
  10. 2023Deep Image Harmonization in Dual Color Spacescited by 12
  11. 2025Comprehensive Relighting: Generalizable and Consistent Monocular Human Relighting and Harmonizationcited by 12
  12. 2022Painterly Image Harmonization in Dual Domainscited by 11
  13. 2022Memory-Driven Text-to-Image Generationcited by 10
  14. 2022Deep Video Harmonization with Color Mapping Consistencycited by 9
  15. 2022FRIH: Fine-grained Region-aware Image Harmonizationcited by 9
  16. 2022Semantic-guided Multi-Mask Image Harmonizationcited by 6
  17. 2023Foreground Object Search by Distilling Composite Image Featurecited by 6
  18. 2023Deep Image Harmonization with Learnable Augmentationcited by 5
  19. 2022GALA: Toward Geometry-and-Lighting-Aware Object Search for Compositingcited by 4
  20. 2025MagicQuillV2: Precise and Interactive Image Editing with Layered Visual Cuescited by 4
  21. 2022Cut-Paste Consistency Learning for Semi-Supervised Lesion Segmentationcited by 3
  22. 2023Image Harmonization with Diffusion Modelcited by 3
  23. 2023Reference-based Painterly Inpainting via Diffusion: Crossing the Wild Reference Domain Gapcited by 3
  24. 2024DiffHarmony: Latent Diffusion Model Meets Image Harmonizationcited by 3
  25. 2024Quality Prediction of AI Generated Images and Videos: Emerging Trends and Opportunitiescited by 3
  26. 2021SIDNet: Learning Shading-aware Illumination Descriptor for Image Harmonizationcited by 2
  27. 2022Interactive Portrait Harmonizationcited by 2
  28. 2022Spatial Transformation for Image Composition via Correspondence Learningcited by 2
  29. 2022gCoRF: Generative Compositional Radiance Fieldscited by 2
  30. 2021A Generative Adversarial Framework for Optimizing Image Matting and Harmonization Simultaneouslycited by 1
  31. 2022Image Harmonization with Region-wise Contrastive Learningcited by 1
  32. 2024Diverse Image Harmonizationcited by 1
  33. 2025POLAR: A Portrait OLAT Dataset and Generative Framework for Illumination-Aware Face Modelingcited by 1
  34. 2021Deep Image Compositingcited by 0
  35. 2021HYouTube: Video Harmonization Datasetcited by 0
  36. 2023Virtual Accessory Try-On via Keypoint Hallucinationcited by 0
  37. 2023Painterly Image Harmonization via Adversarial Residual Learningcited by 0
  38. 2024Harmonizing Attention: Training-free Texture-aware Geometry Transfercited by 0

50 most cited of 52 All 52 (JSON)