Text-Vision Co-Instructed Image Editing

Hugging Face Daily Papers Papers

Summary

A new framework called TV-Edit combines textual instructions and visual prompts for precise image editing, along with a benchmark TV-Edit-Bench for evaluation. The method achieves better spatial control and semantic faithfulness than existing approaches.

Existing image editing methods can be generally categorized into textual instruction-based and visual prompt-based ones. Textual instructions are semantically expressive, but are limited by the coarse granularity of spatial control of the editing results. In contrast, visual prompts such as drag and point can provide precise spatial guidance, but are limited by the inherent ambiguity in semantic intent. To unify the strength of textual and visual prompts, we present Text-Vision Co-Instructed Image Editing, which jointly models textual instructions as semantic intent and sparse visual instructions as spatial guidance, aiming to achieve precise and intent-faithful image manipulation. To this end, we first construct a textual-visual instruction paired dataset with more than 23K samples derived from dynamic videos, enabling aligned supervision for cross-modal instruction. We then propose TV-Edit, a Textual-Visual instruction unified Editing framework to contextualize drag or point-based visual instructions with image-text semantics and lift them into semantic-aware control representations for pretrained editing backbones. By integrating semantic intent and spatial constraints, TV-Edit leads to more precise spatial control, less instruction ambiguity, and stronger structural consistency than text-only or drag-based alternatives. Finally, we establish TV-Edit-Bench, a deliberately designed benchmark to evaluate semantic faithfulness, spatial alignment, and visual consistency with ground-truth references and controlled textual-visual variations for reliable assessment. Our experiments across multiple editing backbones demonstrate that TV-Edit consistently yields more precise and intent-faithful edits, significantly outperforming state-of-the-art instruction-based and drag-based baselines.
Original Article
View Cached Full Text

Cached at: 06/17/26, 11:37 AM

Paper page - Text-Vision Co-Instructed Image Editing

Source: https://huggingface.co/papers/2606.16767

Abstract

A unified text-visual image editing framework is presented that combines semantic intent from textual instructions with spatial guidance from visual prompts to achieve more precise and faithful image manipulation.

Existing image editing methods can be generally categorized intotextual instruction-basedandvisual prompt-basedones. Textual instructions are semantically expressive, but are limited by the coarse granularity ofspatial controlof the editing results. In contrast, visual prompts such as drag and point can provide precisespatial guidance, but are limited by the inherent ambiguity insemantic intent. To unify the strength of textual and visual prompts, we present Text-Vision Co-Instructed Image Editing, which jointly models textual instructions assemantic intentand sparse visual instructions asspatial guidance, aiming to achieve precise and intent-faithful image manipulation. To this end, we first construct atextual-visual instruction paired datasetwith more than 23K samples derived from dynamic videos, enabling aligned supervision forcross-modal instruction. We then proposeTV-Edit, a Textual-Visual instruction unified Editing framework to contextualize drag or point-based visual instructions with image-text semantics and lift them intosemantic-aware control representationsforpretrained editing backbones. By integratingsemantic intentand spatial constraints,TV-Editleads to more precisespatial control, less instruction ambiguity, and strongerstructural consistencythan text-only or drag-based alternatives. Finally, we establishTV-Edit-Bench, a deliberately designed benchmark to evaluate semantic faithfulness, spatial alignment, and visual consistency with ground-truth references and controlled textual-visual variations for reliable assessment. Our experiments across multiple editing backbones demonstrate thatTV-Editconsistently yields more precise andintent-faithful edits, significantly outperforming state-of-the-art instruction-based and drag-based baselines.

View arXiv pageView PDFProject pageGitHub14Add to collection

Get this paper in your agent:

hf papers read 2606\.16767

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2606.16767 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.16767 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.16767 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Uni-Edit: Intelligent Editing Is A General Task For Unified Model Tuning

Hugging Face Daily Papers

Uni-Edit proposes using intelligent image editing as a single general task to simultaneously improve unified multimodal models' understanding, generation, and editing capabilities, with an automated data synthesis pipeline creating complex editing instructions.

TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment

Hugging Face Daily Papers

TIPSv2 introduces enhanced vision-language pretraining techniques including patch-level distillation, an upgraded masked image objective (iBOT++), and improved caption sampling strategies to achieve superior dense patch-text alignment. The resulting family of image-text encoder models demonstrates strong performance across 9 tasks and 20 datasets.