SyncLight: Single-Edit Multi-View Relighting

David Serrano-Lozano; Anand Bhattad; Luis Herranz; Jean-François Lalonde; Javier Vazquez-Corral

cs.CVJan 23, 2026

SyncLight: Single-Edit Multi-View Relighting

Authors: David Serrano-Lozano, Anand Bhattad, Luis Herranz, Jean-François Lalonde, Javier Vazquez-Corral

Organizations: Computer Vision Center Universitat Autònoma de Barcelona · Johns Hopkins University · Universidad Politécnica de Madrid · Université Laval

Abstract

We present SyncLight, a method to enable consistent, parametric control over light sources across multiple uncalibrated views of a static scene conditioned on a single view. While single-view relighting has advanced significantly, existing generative approaches struggle to maintain the rigorous lighting consistency essential for multi-camera broadcasts, stereoscopic cinema, and virtual production. SyncLight addresses this by enabling precise control over light intensity and color across a multi-view capture of a scene, conditioned on a single reference edit. Our method leverages a multi-view diffusion transformer trained using a latent bridge matching formulation, achieving high-fidelity relighting of the entire image set in a single inference step. To facilitate training, we introduce a large-scale hybrid dataset comprising diverse synthetic environments -- curated from existing sources and newly designed scenes -- alongside high-fidelity, real-world multi-view captures under calibrated illumination. Though trained only on image pairs, SyncLight generalizes zero-shot to an arbitrary number of viewpoints, effectively propagating lighting changes across all views, without requiring camera pose information. SyncLight enables practical relighting workflows for multi-view capture systems.

Figures & tables

Figure 1: Multi-view light editing with SyncLight. Given an uncalibrated multi-view capture of a static scene (1), the user picks a reference view , clicks one or more visible light sources (circle markers), and sets their intensity and chromaticity ; the other views never need to be inspected . SyncLight (2) relights all input views, reference included (3), in a single feedforward pass via latent bridge matching. Per-view light control baselines (bottom-left) hallucinate, lack precise control, and are inconsistent across views. SyncLight is geometrically consistent, needs no camera poses, generalizes to any number of views, and is 10 $\times$ faster than current models.

Figure 2: SyncLight formulates multi-view relighting as a conditional flow matching problem in latent space. (Left) Input scenes under source ( $\mathbf{x}_{\text{src}}$ ) and target lighting ground truth ( $\mathbf{x}_{\text{tar}}$ ) are encoded into latents ( $\mathbf{z}_{\text{src}},\mathbf{z}_{\text{tar}}$ ) via a VAE encoder. This is done for both the “Reference” (0) and “Other” (1) views. (Middle) During training, we sample a timestep $t$ and construct a Bridge Matching (BM) interpolant $\mathbf{z}_{t}$ . Our backbone, “Multi-View SD”, is conditioned on a user-specified “Lightmap” (encoding color as intensity and chromaticity) derived from the “Reference view”. To enforce consistency, the backbone processes both views simultaneously. (Right) The model predicts the target latents, which are decoded into relit images ( $\hat{\mathbf{x}}_{T}$ ). The network is optimized using a hybrid objective $\mathcal{L}$ combining latent flow matching loss ( $\mathcal{L}_{\text{lbm}}$ ) with pixel-level reconstruction losses ( $\mathcal{L}_{\text{pix}}$ ) for each view to ensure high-fidelity, consistent relighting.

	Method	PSNR	SSIM	$\Delta E_{00}$	LPIPS	PSNR	SSIM	$\Delta E_{00}$	LPIPS	PSNR	SSIM	$\Delta E_{00}$	LPIPS	VGGTm	Time (s)
		Infinigen				BlenderKit				Real captures
Ref. view	ScribbleLight	10.89	.518	30.67	.382	11.02	.528	29.26	.372	11.78	.483	28.36	.375	-	58.2 $\pm$ 1.14
	Flux.2-dev	14.23	.672	20.31	.338	16.71	.704	17.22	.291	18.32	.757	14.28	.264	-	55.6 $\pm$ 1.02
	LightLab*	29.86	.941	3.08	.131	26.38	.907	4.92	.147	28.42	.880	4.61	.199	-	1.17 $\pm$ 0.02
	GR3EN	25.92	.893	4.38	.158	21.74	.803	8.93	.173	22.31	.782	9.31	.238	-	72.3 $\pm$ 3.51
	SyncLight	31.32	.950	2.47	.119	27.16	.915	4.02	.134	30.34	.895	3.73	.196	-	-

Table 1: Quantitative multi-view image relighting results on our SyncLight test set. Each section of the table indicates which view is used to compute metrics: the “Reference (Ref.) view” (where edits are specified), “Second (Sec.) view” (another view in the set), and “Additional (Add.) views” (all but the reference view). Best results, achieved by SyncLight in all cases, are highlighted. See text for details on baselines. As SyncLight is always run on more than one image; no inference time is reported for the “Ref. view.” VGGTm (%) measures cross-view consistency against the ground truth (see text); it is undefined for the reference view alone.

Figure 3: Qualitative results showcasing SyncLight’s controllability and generalization. Top-left: precise control over light color, varying the chromaticity of a table lamp across three target hues. Top-right: continuous control over light intensity, from off ( $0.0$ ) to fully on ( $1.0$ ). Bottom-left: robustness to large camera angle changes, with input views rotated by up to $180^{\circ}$ relative to the reference. Bottom-right: generalization across diverse scenes and unseen light types. SyncLight produces consistent relighting across all viewpoints.

Figure 4: Results in the Real Split from our dataset. SyncLight relights the images correctly. Flux.2-dev struggles with cross-view consistency, producing visually plausible but geometrically inconsistent results (e.g., the window in the second row). Meanwhile, the LightLab*+LumiNet baseline maintains better consistency by explicitly relighting each view, but fails to propagate lighting effects.

Figure 6: Relighting a scene with seven different views. Top row: Input images. Bottom row: SyncLight results. Note how all the views are consistently modified. We want to emphasize views 2 and 4, in which the lamp turned on is not visible, yet the effects in the scene match those of the other views.

Figure 7: Qualitative results on out-of-distribution images from the RealEstate10K dataset Google Research (2018) . SyncLight outperforms all the other methods, including a per-view informed version of LightLab*, demonstrating that the joint processing of all views is beneficial even for single-light relighting.

Appendix figures & tables15 assets

Supplementary material from the paper’s appendix.

Appendix

Panel of Figure 8 — Figure 8: Examples from the SyncLight dataset. Samples were selected randomly with random light colors.

Table 10

Table 11

Figure 11: Multi-view relighting examples when training SyncLight without one of the splits of the dataset at a time.

Figure 12: Instructions given to the users for the User Study.

Figure 13: User study results.

Panel of Figure 14 — Figure 14: SyncLight applied to: (a) video relighting, and (b) novel-view synthesis with radiance fields. Please see the supplementary video on our webpage.

Panel of Figure 15 — Figure 15: Qualitative results on the SyncLight dataset on the Infinigen (1–2), and BlenderKit (3–4) splits. Real split is shown in the main paper.

Panel of Figure 16 — Figure 16: Inference time (left) and peak memory (right) of SyncLight as a function of the number of views $N$ on an NVIDIA A40 (BlenderKit, 32 rendered views per scene). Memory grows linearly; compute grows quadratically.

$N$	PSNR	SSIM	$\Delta E_{00}$	LPIPS	VGGTm	Inf. time (s)
2	30.23	.897	3.56	.197	95.8	1.58 $\pm$ 0.02
4	30.40	.904	3.55	.194	94.8	2.17 $\pm$ 0.02
6	30.45	.907	3.56	.194	94.9	3.01 $\pm$ 0.03
7	30.47	.906	3.56	.192	95.3	3.83 $\pm$ 0.03

Appendix

Table 9: Scaling with the number of views

N

on the Real test split.

Method	Interaction per edited light	Cost per additional view
SyncLight	one click + color + intensity	none
LightLab [ 49 ]	segmentation mask + color + intensity	one mask per view
GR3EN [ 63 ]	segmentation mask + color + intensity	one mask per view + camera poses
LuxRemix [ 42 ]	OLAT harmonization + color + intensity	Plücker embeddings + harmonization

Appendix

Table 10: User input required by each method.

Weak edits ( $\|\Delta L\|<0.5$ )
	Ours vs. GT		$\Delta E_{00}$ vs. input
Region	PSNR	$\Delta E_{00}$	GT	Ours
0–1 $r$ (marker)	30.6	3.12	7.9	8.2
1–2 $r$	30.9	2.84	3.1	3.3
2–4 $r$	31.8	2.62	1.6	1.7
$>$ 4 $r$	32.1	2.71	0.9	1.0

Appendix

Table 11: Local versus global changes on the Real split. Weak and strong edits are separated by the requested lightness change

|\Delta L|

; pixels are grouped by their distance to the edited source in multiples of the marker radius

r

. The last row reports all edits over the full image, as in table 1 .

Figure 17: Additional results on the RealEstate10K dataset.

Figure 18: Additional results on the RealEstate10K dataset

Explore similar work

RelightFormer: Feed-forward Generative Transformer for Multiview Object Relighting
Sep 7, 2026Hejun Wang, Jinxi Li, Junwei Jiang +4Image Relighting Inverse Rendering
Shenzhen Research Institute, The Hong Kong Polytechnic University, Hong Kong
Decoupled Illumination Priors for Spatially Controllable Multi-View Indoor Scene Relighting
Jul 9, 2026Chenjian Gao, Linning Xu, Tianfan XueImage Relighting Visual Priors
Multimedia Laboratory, The Chinese University of Hong Kong · Shanghai AI Laboratory · CPII under InnoHK
LiveLight: Real-time Streaming Video Relighting with Interactive Control
Aug 3, 2026Yue Ma, Jiangming Wang, Yucheng Wang +8Image Relighting Video Diffusion Models
1HKUST · University of Macau · 3THU +2

Sep 7, 2026, cs.CVJ/K move · Enter open · S save

RelightFormer: Feed-forward Generative Transformer for Multiview Object Relighting

Hejun Wang, Jinxi Li, Junwei Jiang, Shiwei Mao+3

Shenzhen Research Institute, The Hong Kong Polytechnic University, Hong Kong

Image Relighting Inverse Rendering Multi-View Multi-Reference Image Generation Objects

Open paper AskPDF arXiv

Image relighting is traditionally tackled via complex inverse rendering pipelines, which suffer from ill-posed optimization, or single-image generative models that ignore crucial multi-view cues necessary for understanding 3D geometry and material interactions. To address these limitations, we introduce a feed-forward generative Transformer for direct single- and multi-view image relighting that entirely bypasses explicit intrinsic property estimation. Adapted from a video foundation model, our architecture features a latent illumination module that dynamically injects target environment maps into spatial features via cross-attention. Furthermore, we employ permutation-invariant positional encodings to symmetrically process unordered multi-view inputs without sequential bias. To train this robust data-driven model, we construct the massive Laval Objaverse Dataset (LOD), comprising 90K objects and 39K unique illuminations. Extensive experiments demonstrate state-of-the-art visual quality, photorealistic relighting quality, and strong zero-shot generalization across single-view, multi-view, and novel-view relighting tasks.

Abstract

Explore similar work

RelightFormer: Feed-forward Generative Transformer for Multiview Object Relighting

Decoupled Illumination Priors for Spatially Controllable Multi-View Indoor Scene Relighting

LiveLight: Real-time Streaming Video Relighting with Interactive Control