{"paper":{"title":"SplatWeaver: Learning to Allocate Gaussian Primitives for Generalizable Novel View Synthesis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"SplatWeaver learns to assign different numbers of 3D Gaussians to different scene regions for better feed-forward novel view synthesis.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fan Li, Mingwen Shao, Wangmeng Zuo, Yecong Wan","submitted_at":"2026-05-08T05:51:34Z","abstract_excerpt":"Generalizable novel view synthesis aims to render unseen views from uncalibrated input images without requiring per-scene optimization. Recent feed-forward approaches based on 3D Gaussian Splatting have achieved promising efficiency and rendering quality. However, most of them assign a fixed number of Gaussians to each pixel or voxel, ignoring the spatially varying complexity of real-world scenes. Such uniform allocation often wastes Gaussian primitives in smooth regions while providing insufficient capacity for fine structures, complex geometry, and high-frequency details. This motivates us t"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"SplatWeaver consistently outperforms state-of-the-art methods, delivering more faithful novel-view renderings with fewer Gaussian primitives.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the combination of cardinality experts, pixel-level routing, high-frequency prior, and routing regularization will stably produce complexity-aware allocations that generalize across scenes without overfitting or requiring per-scene optimization.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SplatWeaver dynamically allocates Gaussian primitives via cardinality experts and pixel-level routing guided by high-frequency cues for improved generalizable novel view synthesis.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SplatWeaver learns to assign different numbers of 3D Gaussians to different scene regions for better feed-forward novel view synthesis.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"39d6155cb5e0ab4450314b696016fb84a1987c90f3c5770c036da6cb922f0a10"},"source":{"id":"2605.07287","kind":"arxiv","version":2},"verdict":{"id":"64fd980a-1b9e-4907-865b-785368e4ff1d","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-11T01:12:03.968368Z","strongest_claim":"SplatWeaver consistently outperforms state-of-the-art methods, delivering more faithful novel-view renderings with fewer Gaussian primitives.","one_line_summary":"SplatWeaver dynamically allocates Gaussian primitives via cardinality experts and pixel-level routing guided by high-frequency cues for improved generalizable novel view synthesis.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the combination of cardinality experts, pixel-level routing, high-frequency prior, and routing regularization will stably produce complexity-aware allocations that generalize across scenes without overfitting or requiring per-scene optimization.","pith_extraction_headline":"SplatWeaver learns to assign different numbers of 3D Gaussians to different scene regions for better feed-forward novel view synthesis."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.07287/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T11:02:03.315357Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-20T06:33:58.802064Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T17:01:19.057851Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T11:54:25.983616Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"16bea75fc8b89514c68d3e3c3718b73a9d9ec832c9527d8d3c4a7f6ec66c221e"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"f0a8e9e18f95d0108e26aa1d950830f8f650c6ef8cb808c44f0d2d702b8a0fa6"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}