Research map / Detection and segmentation / Open-set and few-shot detection
Alterbute: Editing Intrinsic Attributes of Objects in Images
Tal Reiss, Daniel Winter, Matan Yaakov Cohen, Alex Rav-Acha, Yael Pritch, Ariel Shamir, 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 |
|---|---|---|---|
| train for 100K steps instead of 50K (ablation) | Alterbute · VLM preference vs baselines (Gemini) | preference rate (%): 78.0 -> 86.1 | Improved |
| train for 100K steps instead of 50K (ablation) | Alterbute · VLM preference vs baselines (Claude) | preference rate (%): 76.3 -> 84.9 | Improved |
| add grid (1x2 cross-image self-attention) identity conditioning instead of channel-wise concatenation (ablation) | Alterbute · Intrinsic attribute editing | edit success (qualitative) | Improved |
| define identity by VNE cluster instead of DINOv2 features (ablation) | Alterbute · Intrinsic attribute editing | edit quality (qualitative) | Improved |
| define identity by VNE cluster instead of instance-retrieval features (ablation) | Alterbute · Intrinsic attribute editing | edit quality (qualitative) | Improved |
| train with relaxed VNE reference instead of in-place editing (target as its own reference) (ablation) | Alterbute · Intrinsic attribute editing | edit quality (qualitative) | Improved |
| Alterbute versus FluxKontext | Alterbute · VLM preference (Gemini) | preference rate (%): 61.2 | Improved |
| Alterbute versus Qwen-image-editing | Alterbute · VLM preference (Gemini) | preference rate (%): 49.8 | No clear effect |
| Alterbute versus UltraEdit on conventional DINO metric | Alterbute · Conventional protocol | DINO: 0.841 -> 0.815 | Hurt |
| Alterbute versus Diptych on CLIP-T | Alterbute · Conventional protocol | CLIP-T: 0.313 -> 0.321 | Improved |
Closest papers by meaning
- A²-Edit: Precise Reference-Guided Image Editing of Arbitrary Objects and Ambiguous Masks (ECCV 2026 · similarity 0.87)
- Paint by Inpaint: Learning to Add Image Objects by Removing Them First (CVPR 2025 · similarity 0.85)
- PAIR Diffusion: A Comprehensive Multimodal Object-Level Image Editor (CVPR 2024 · similarity 0.85)
- Benchmarking Segmentation Models with Mask-Preserved Attribute Editing (CVPR 2024 · similarity 0.85)
- IMPRINT: Generative Object Compositing by Learning Identity-Preserving Representation (CVPR 2024 · similarity 0.85)
- Localizing and Editing Knowledge In Text-to-Image Generative Models (ICLR 2024 · similarity 0.84)
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.
