{"paper":{"title":"SyncLight: Single-Edit Multi-View Relighting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"SyncLight lets users edit the lighting in one view and automatically applies consistent changes to all other views of the scene in a single step.","cross_cats":["cs.GR"],"primary_cat":"cs.CV","authors_text":"Anand Bhattad, David Serrano-Lozano, Javier Vazquez-Corral, Jean-Fran\\c{c}ois Lalonde, Luis Herranz","submitted_at":"2026-01-23T18:59:57Z","abstract_excerpt":"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 d"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"SyncLight enables precise control over light intensity and color across a multi-view capture of a scene, conditioned on a single reference edit, achieving high-fidelity relighting of the entire image set in a single inference step while generalizing zero-shot to an arbitrary number of viewpoints without requiring camera pose information.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That training exclusively on image pairs with a latent bridge matching formulation is sufficient to guarantee lighting consistency and zero-shot generalization to arbitrary numbers of uncalibrated views in real-world static scenes.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SyncLight propagates a single reference lighting edit consistently across arbitrary numbers of uncalibrated views using a multi-view diffusion transformer trained with latent bridge matching on a hybrid synthetic-real dataset.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SyncLight lets users edit the lighting in one view and automatically applies consistent changes to all other views of the scene in a single step.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"1899b43123f5054252ee8b7074f7c05c2fa6c880231c850102ac4adac0ec32d2"},"source":{"id":"2601.16981","kind":"arxiv","version":2},"verdict":{"id":"f5f02e9f-d47a-44c4-9a1c-958013b19092","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T11:38:54.744076Z","strongest_claim":"SyncLight enables precise control over light intensity and color across a multi-view capture of a scene, conditioned on a single reference edit, achieving high-fidelity relighting of the entire image set in a single inference step while generalizing zero-shot to an arbitrary number of viewpoints without requiring camera pose information.","one_line_summary":"SyncLight propagates a single reference lighting edit consistently across arbitrary numbers of uncalibrated views using a multi-view diffusion transformer trained with latent bridge matching on a hybrid synthetic-real dataset.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That training exclusively on image pairs with a latent bridge matching formulation is sufficient to guarantee lighting consistency and zero-shot generalization to arbitrary numbers of uncalibrated views in real-world static scenes.","pith_extraction_headline":"SyncLight lets users edit the lighting in one view and automatically applies consistent changes to all other views of the scene in a single step."},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}