Research map / Image generation / Autoregressive and latent generation
ObjectStitch: Object Compositing With Diffusion Model
Yizhi Song, Zhifei Zhang, Zhe Lin, Scott Cohen, Brian Price, Jianming Zhang, Soo Ye Kim, Daniel Aliaga
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add the content adaptor instead of removing it (ablation) | ObjectStitch · Real-world evaluation dataset | FID: 18.8254 -> 15.4331 | Improved |
| Add appearance-preservation fine-tuning of the content adaptor (ablation) | ObjectStitch · Real-world evaluation dataset | CLIP image score: 96.8125 -> 97.0625 | Improved |
| Add crop and shift augmentation during training (ablation) | ObjectStitch · Real-world evaluation dataset | FID: 18.3698 -> 15.4331 | Improved |
| Use the content adaptor instead of BLIP-generated text embeddings (ablation) | ObjectStitch · Real-world evaluation dataset | CLIP image score: 95.8125 -> 97.0625 | Improved |
Later papers that compared with this method
6 later accepted papers report ObjectStitch as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.
- CareCom: Generative Image Composition with Calibrated Reference Features (AAAI 2026 · better on 6 of 6 shared results)
- UniCombine: Unified Multi-Conditional Combination with Diffusion Transformer (ICCV 2025 · better on 2 of 2 shared results)
- PixPerfect: Seamless Latent Diffusion Local Editing with Discriminative Pixel-Space Refinement (NeurIPS 2025 · better on 1 of 1 shared result)
- Refine-by-Align: Reference-Guided Artifacts Refinement through Semantic Alignment (ICLR 2025 · better on 1 of 1 shared result)
- ObjectMate: A Recurrence Prior for Object Insertion and Subject-Driven Generation (ICCV 2025 · better on 1 of 1 shared result)
- IMPRINT: Generative Object Compositing by Learning Identity-Preserving Representation (CVPR 2024 · better on 7 of 7 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- IMPRINT: Generative Object Compositing by Learning Identity-Preserving Representation (CVPR 2024 · similarity 0.87)
- $\textit{Bifr\"ost}$: 3D-Aware Image Compositing with Language Instructions (NeurIPS 2024 · similarity 0.87)
- MV-CoLight: Efficient Object Compositing with Consistent Lighting and Shadow Generation (NeurIPS 2025 · similarity 0.86)
- Multitwine: Multi-Object Compositing with Text and Layout Control (CVPR 2025 · similarity 0.86)
- PAIR Diffusion: A Comprehensive Multimodal Object-Level Image Editor (CVPR 2024 · similarity 0.86)
- Paint by Inpaint: Learning to Add Image Objects by Removing Them First (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
