Research map / Image generation / Text-to-image diffusion
ObjectMate: A Recurrence Prior for Object Insertion and Subject-Driven Generation
Daniel Winter, Asaf Shul, Matan Cohen, Dana Berman, Yael Pritch, Alex Rav-Acha, Yedid Hoshen
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 2 more reference views (3 instead of 1) (ablation) | ObjectMate · Object insertion | IR: 0.803 -> 0.858 | Improved |
| add 2 more reference views (ablation) | ObjectMate · Object insertion | DINO: 0.868 -> 0.885 | Improved |
| add 2 more reference views (ablation) | ObjectMate · DreamBench subject generation | IR: 0.739 -> 0.750 | Improved |
| add 2 more reference views (ablation) | ObjectMate · DreamBench subject generation | CLIP-I: 0.770 -> 0.773 | No clear effect |
| use IR retrieval features instead of DINO for building the dataset (ablation) | ObjectMate · Object insertion user study | identity preference: 63% | Improved |
| scale up unsupervised dataset size (ablation) | ObjectMate · Object insertion | identity and DINO metrics | Improved |
| use public Open Images data and public IR features (ablation) | ObjectMate · Object insertion | IR / DINO | No clear effect |
| use IR for identity metric instead of CLIP-I | user preference prediction · Subject generation | prediction accuracy: 64.7% -> 72.9% | Improved |
| use IR for identity metric instead of DINO | user preference prediction · Object insertion | prediction accuracy: 71.8% -> 79.5% | Improved |
| ObjectMate versus ObjectDrop copy-paste | ObjectMate · Object insertion user study | identity preservation: 71% | Hurt |
| ObjectMate versus ObjectDrop | ObjectMate · Object insertion user study | composition: 76% | Improved |
| ObjectMate versus DisenBooth | ObjectMate · DreamBench subject generation | DINO: 0.625 vs 0.607 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- DisenBooth — DisenBooth: Identity-Preserving Disentangled Tuning for Subject-Driven Text-to-Image Generation (ICLR 2024) · better on 2 of 4 shared results
- ObjectStitch — ObjectStitch: Object Compositing With Diffusion Model (CVPR 2023) · better on 1 of 1 shared result
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
- ObjectMover: Generative Object Movement with Video Prior (CVPR 2025 · similarity 0.87)
- Learning to Generate Multiple Objects from Dense and Occluded Layouts (NeurIPS 2026 · similarity 0.87)
- $\textit{Bifr\"ost}$: 3D-Aware Image Compositing with Language Instructions (NeurIPS 2024 · similarity 0.87)
- Add-it: Training-Free Object Insertion in Images With Pretrained Diffusion Models (ICLR 2025 · similarity 0.86)
- Neural Assets: 3D-Aware Multi-Object Scene Synthesis with Image Diffusion Models (NeurIPS 2024 · similarity 0.86)
- DORSal: Diffusion for Object-centric Representations of Scenes $\textit{et al.}$ (ICLR 2024 · 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.
