paper-with-me

Papers

Open-Source Image Editing Models Are Zero-Shot Vision Learners

2026-05-06 · Wei Liu, Jiaxin Lin, Rui Chen arxiv

Recent studies have shown that large generative models can solve vision tasks they were not explicitly trained for. However, existing evidence relies on closed-source models~(Veo~3, Nano Banana Pro) or requires task-specific instruction tuning, leaving open whether publicly available image-editing models possess zero-shot vision abilities out of the box. We conduct a systematic evaluation of three open-source image-editing models -- Qwen-Image-Edit, FireRed-Image-Edit, and LongCat-Image-Edit -- on dense visual prediction tasks \emph{without any fine-tuning}. We benchmark monocular depth estimation on NYUv2 and DIODE, surface normal estimation on NYUv2, and semantic segmentation on Cityscapes, covering both geometric and semantic scene understanding. Results show that open-source image-editing models exhibit non-trivial zero-shot visual understanding. On NYUv2 surface normals, FireRed-Image-Edit achieves a mean angular error of $17.69^\circ$, surpassing the fine-tuned Marigold ($20.86^\circ$) and matching the instruction-tuned Vision Banana ($17.78^\circ$) without any task-specific training. On NYUv2 depth estimation, LongCat-Image-Edit obtains $δ_1{=}0.822$ with affine alignment, and Qwen-Image-Edit leads on DIODE Indoor ($δ_1{=}0.868$). On Cityscapes semantic segmentation, Qwen-Image-Edit reaches 25.7 mIoU at the 19-class level and 49.5 mIoU at a coarser 7-category level. By comparing three independently trained editors, we test whether zero-shot vision ability is an emergent property of image-editing pretraining rather than a model-specific artifact. Code, evaluation scripts, and all results are publicly released to serve as a reproducible baseline for future work.

📄 PDF Abstract BibTeX arXiv:2605.04566

Code (0)

등록된 구현이 없습니다.

Tasks

Monocular Depth EstimationSemantic SegmentationScene UnderstandingImage Editing

Results from the Paper

RankTaskDatasetModelMetrics
#21 Semantic Segmentation Cityscapes Open-Source Image Editing Models Are Zer mIoU: 0.868

Similar Papers 제목 키워드 기반

The Curious Case of End Token: A Zero-Shot Disentangled Image Editing using CLIP

2024-06-01 · Hidir Yesiltepe, Yusuf Dalva, Pinar Yanardag

Diffusion models have become prominent in creating high-quality images. However, unlike GAN models celebrated for their ability to edit images in a disentangled manner, diffusion-based text-to-image models struggle to ac…

AttributeVideo Editing

FateZero: Fusing Attentions for Zero-shot Text-based Video Editing

2023-03-16 · ICCV 2023 1 · Chenyang Qi, Xiaodong Cun, Yong Zhang, Chenyang Lei 외

The diffusion-based generative models have achieved remarkable success in text-based image generation. However, since it contains enormous randomness in generation progress, it is still challenging to apply such models f…

AttributeText-to-Video EditingVideo EditingVideo Style Transfer

Zero-Shot Video Editing Using Off-The-Shelf Image Diffusion Models

2023-03-30 · Wen Wang, Yan Jiang, Kangyang Xie, Zide Liu 외

Large-scale text-to-image diffusion models achieve unprecedented success in image generation and editing. However, how to extend such success to video editing is unclear. Recent initial attempts at video editing require …

Video AlignmentVideo Editing

Zero-Shot Unsupervised and Text-Based Audio Editing Using DDPM Inversion

2024-02-15 · Hila Manor, Tomer Michaeli

Editing signals using large pre-trained models, in a zero-shot manner, has recently seen rapid advancements in the image domain. However, this wave has yet to reach the audio domain. In this paper, we explore two zero-sh…

NEP: Autoregressive Image Editing via Next Editing Token Prediction

2025-08-08 · Huimin Wu, Xiaojian Ma, Haozhe Zhao, Yanpeng Zhao 외 arxiv

Text-guided image editing involves modifying a source image based on a language instruction and, typically, requires changes to only small local regions. However, existing approaches generate the entire target image rath…

Image GenerationImage Editing