LOCI: Localized Continuous Control for Multi-Subject Image Editing

Sai Vidyaranya Nuthalapati; Shlok Kumar Mishra

cs.CVOct 22, 2026

LOCI: Localized Continuous Control for Multi-Subject Image Editing

Authors: Sai Vidyaranya Nuthalapati, Shlok Kumar Mishra

Organizations: Meta AI

Abstract

Recent instruction-following image editors such as Flux.1-Kontext enable text-driven editing but lack fine-grained control over where and how much an edit is applied. Existing approaches provide either continuous but global control or spatially localized editing at a fixed strength. To our knowledge, none jointly supports explicit spatial localization, continuous per-region strength, and independent simultaneous control of multiple subjects. We present LOCI (LOcalized Continuous control for multi- subject Image editing). Given a multi-subject image with per-subject regions and instructions, LOCI edits each subject at an independently chosen, continuously adjustable strength, for instance gradually aging one person while adding a beard to another with minimal cross-region leakage. LOCI pairs subject-level attention binding for spatial isolation with a Regional LoRA layer that replaces the LoRA's scalar coefficient with a per-token scaling vector aligned to the spatial layout. A single adapter is trained at one suppression point, and continuous edit strengths follow by rescaling its regional residual at inference. Experiments on multi-subject scenes demonstrate that LOCI achieves strong edit fidelity with significantly reduced cross-region interference compared to baselines, while maintaining continuous control over edit strength

Figures & tables

Panel of Figure 1 — Figure 1: LOCI gives every subject its own continuous edit control. From one input, the top row strengthens the edit on the first teapot while the second stays pinned at the neutral anchor, the middle row does the converse, and the bottom row moves both controls together. The boxes on the input are the spatial annotations supplied to the model: red for subject 1 and blue for subject 2, matching the row labels.

Figure 2: Architecture of LOCI. Stage 1 : the user supplies a source image, per-region bounding boxes $B_{n}$ , edit instructions, and independent scalar controls $\alpha_{n}$ . Stage 2 : text and image tokens are organized into the partition $\{T_{n},L_{n},C_{n}\}$ indexed by subject, with trainable low-rank adapters ( $\Delta W$ , marked with ) whose residuals are gated per token by $\alpha_{n}$ ; background tokens $L_{u},C_{u}$ receive zero adapter contribution. Stage 3 : the frozen Flux.1-Kontext transformer processes the sequence through 19 dual-stream and 38 single-stream blocks under a three-zone attention schedule, with harmonization in the outer blocks (0–9, 47–56) and disentanglement in the middle blocks (10–46). Inset : the query–key attention matrices for both regimes. Under disentanglement, each subject’s tokens attend only within their own partition; under harmonization, subject image tokens additionally read all visual tokens (dashed cells), restoring the cross-region evidence needed for background coherence.

Panel of Figure 3 — Figure 3: Independent and joint control across four scenes. Each block shows its input and instructions on the left, with red and blue boxes marking the spatial annotations given to the model. The three rows sweep subject 1, subject 2, and both, and within a row the edit strengthens from left to right. see section 4.3 .

Protocol	Method	CLIP	SigLIP	CLIP	SigLIP	CLIP	SigLIP	LPIPS-Alex	LPIPS-VGG	DINOv2	Identity
		Extrapolation $\uparrow$		Direction corr. $\downarrow$		Mean range $\uparrow$		Disentanglement $\downarrow$
A	(1) SliderEdit	0.2431	0.1114	$-0.4907$	$-0.4666$	0.0575	0.0470	0.3315	0.3826	0.1612	0.3640
A	(2) SliderEdit $+$ IDAttn	0.2500	0.1187	$-0.6326$	$-0.6192$	0.0726	0.0587	0.1508	0.2010	0.0609	0.2598
A	(4) Regional LoRA without IDAttn	0.2445	0.1133	$-0.5829$	$-0.5689$	0.0761	0.0625	0.3085	0.3581	0.1700	0.3406
A	(5) LOCI (ours)	0.2527	0.1207	$\mathbf{-0.7081}$	$\mathbf{-0.6953}$	0.0894	0.0721	0.1467	0.1979	0.0550	0.2548
B	(1) SliderEdit	0.2426	0.1121	$-0.5503$	$-0.5106$	0.0589	0.0488	0.3346	0.3875	0.1471	0.3502

Table 1: Protocols A and B. Best per column bold , second best underlined , ranked within each protocol block.

	Protocol C: joint editing		Protocol D: two-dimensional control
Method	Regional success $\uparrow$	Worst-region $\Delta$ CLIP $\uparrow$	Intended response $\uparrow$	Semantic cross-talk $\downarrow$	LPIPS cross-talk $\downarrow$
(1) SliderEdit	0.8231	0.0045	0.0307	0.6456	0.6597
(2) SliderEdit $+$ IDAttn	0.8773	0.0122	0.0353	0.2776	0.1909
(3) IDAttn only	0.9153	0.0282	not slider-capable
(4) Regional LoRA without IDAttn	0.7577	$-0.0015$	0.0393	0.7146	0.9599
(5) LOCI (ours)	0.9099	0.0251	0.0458	0.2211	0.1860

Table 2: Protocols C and D. Protocol D applies only to slider-capable methods. Best per column bold , second best underlined .

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Panel of Figure 4 — Figure 4: Four independent controls in one scene. The input carries four spatial annotations, drawn in red , blue , green and orange , each with its own instruction. Each of the first four rows moves one chair’s control while the other three are held at the neutral anchor $\alpha=1$ ; the last row moves all four together. Every column shows an edited output; the unedited scene is the input on the left.

Panel of Figure 5 — Figure 5: Four independent controls on a second scene. Each of the first four rows moves one vase’s control while the other three are held at the neutral anchor $\alpha=1$ , and the last row moves all four together. The first column is the unedited output at $\alpha=1$ .

Image Editing

Image editing changes selected visual content while preserving the parts of an image that should remain unchanged.

See all 313 papers

Rethinking Where to Edit: Task-Aware Localization for Instruction-Based Image Editing
Apr 22, 2026Jingxuan He, Xiyu Wang, Mengyu Zheng +3Text-Guided Image Editing Image Editing
The University of Sydney Sydney, Australia
Scalable Minimal-Change Learning for Controllable Image Editing
Oct 5, 2026Shuo Chen, Fengming Huang, Yu Yao +2Text-Guided Image Editing Image Editing
Sydney AI Centre, The University of Sydney
VeloEdit: Training-Free Consistent and Continuous Instruction-Based Image Editing via Velocity Field Decomposition
Mar 11, 2026Zongqing Li, Zhihui Liu, Yujie Xie +3Training-Free Image Editing Image Editing
Xiamen University · Truesight

Apr 22, 2026J/K move · Enter open · S save

Rethinking Where to Edit: Task-Aware Localization for Instruction-Based Image Editing

Jingxuan He, Xiyu Wang, Mengyu Zheng, Xiangyu Zeng+2

The University of Sydney Sydney, Australia

Text-Guided Image Editing Image Editing

Open paper AskPDF arXiv

Instruction-based image editing (IIE) aims to modify an image according to a natural language instruction. Despite recent advances in diffusion transformers, existing methods often introduce unintended changes to regions unrelated to the requested edit. We attribute this limitation to the absence of an explicit mechanism for edit localization. Different editing operations (e.g., subject addition, removal, and replacement) induce distinct spatial patterns, yet existing IIE models typically perform localization in a task-agnostic manner. To address this limitation, we propose a training-free, task-aware edit localization framework that exploits the intrinsic source and target image streams of IIE models. For each image stream, we construct feature centroids from attention-based edit cues, and then partition tokens into edit and non-edit regions based on feature similarity. Observing that effective localization is inherently task-dependent, we introduce a unified mask construction strategy that selectively leverages the source and target streams according to the editing task. We also provide a systematic analysis of our underlying insights and design choices. Extensive experiments on EdiVal-Bench demonstrate that our framework consistently improves content consistency in non-edit regions while maintaining strong instruction-following performance on top of Step1X-Edit and Qwen-Image-Edit.

Abstract

Image Editing

Explore similar work

Rethinking Where to Edit: Task-Aware Localization for Instruction-Based Image Editing

Scalable Minimal-Change Learning for Controllable Image Editing

VeloEdit: Training-Free Consistent and Continuous Instruction-Based Image Editing via Velocity Field Decomposition