{"rewrite":{"id":"r_caedd72d44edcd0690248b52","clusterId":"c_8c6dacb676bad20cc9da76dd","slug":"nvidia-develops-artifixer-ai-model-for-3d-scene-generation-from-photos","model":"deepseek-v4-flash","headline":"Nvidia Develops ArtiFixer AI Model for 3D Scene Generation From Photos","summary":"Nvidia has announced ArtiFixer, an AI model based on the Wan 2.1 video generation AI that can generate 3D scenes from multiple photos. The model fills in missing parts using generative processing, improving on conventional 3D Gaussian Splatting methods that often fail to maintain scene consistency. ArtiFixer comes in three variants: ArtiFixer, ArtiFixer3D, and ArtiFixer3D+.","whyItMatters":"ArtiFixer addresses a known weakness in 3D Gaussian Splatting by generating and inserting parts not captured in reference photos, producing sharper and more consistent 3D scenes.","webCardHtml":"\u003cp\u003eNvidia has developed ArtiFixer, an AI model that generates 3D scenes from multiple photos by filling in missing parts using generative processing. The model is based on the Wan 2.1 video generation AI and has approximately 16.9 billion parameters. It was trained in two stages: first to generate and insert parts not in the photos, then distilled into an autoregressive model that generates hundreds of frames from a single frame.\u003c/p\u003e\u003cp\u003eArtiFixer comes in three variants. ArtiFixer uses an autoregressive model to generate views. ArtiFixer3D distills that output into a 3D representation. ArtiFixer3D+ applies an autoregressive model as post-processing to ArtiFixer3D\u0026#39;s results. Nvidia says ArtiFixer3D+ can generate scenes that are both sharp and highly consistent, outperforming conventional methods like 3DGUT, GenFusion, and GSFixer in quality.\u003c/p\u003e","blueskyPost":"Nvidia announced ArtiFixer, an AI model that generates 3D scenes from multiple photos by filling in missing parts. Based on Wan 2.1, it comes in three variants: ArtiFixer, ArtiFixer3D, and ArtiFixer3D+.","twitterPost":"Nvidia announced ArtiFixer, an AI model that generates 3D scenes from multiple photos by filling in missing parts. Based on Wan 2.1, it comes in three variants: ArtiFixer, ArtiFixer3D, and ArtiFixer3D+.","threadsPost":null,"newsletterBlurb":"Nvidia has announced ArtiFixer, an AI model that generates 3D scenes from multiple photos by filling in missing parts using generative processing. Based on the Wan 2.1 video generation AI, it comes in three variants: ArtiFixer, ArtiFixer3D, and ArtiFixer3D+.","attributionJson":"[{\"source\":\"GIGAZINE\",\"url\":\"https://gigazine.net/news/20260623-nvidia-artifixer-3d-representation/\",\"title\":\"NVIDIA Announces AI Model 'ArtiFixer' That Generates 3D Scenes from Multiple Photos, Can Also Fill in Missing Parts via Generative Processing\"}]","lintFlagsJson":null,"lintHits":0,"costUsd":0,"inputTokens":4309,"outputTokens":688,"status":"published","repairAttempts":0,"nextRepairAt":null,"factsAttemptedAt":1782200670,"createdAt":"2026-06-23T07:37:03.000Z","publishedAt":"2026-06-23T07:40:24.000Z","updatedAt":"2026-06-23T07:40:24.000Z"},"cluster":{"id":"c_8c6dacb676bad20cc9da76dd","canonicalTitle":"複数の写真から3Dシーンを生成するAIモデル「ArtiFixer」をNVIDIAが発表、写真にない部分も生成処理で補完可能","representativeArticleId":"a_ff0f08882633447e76324ab7","sourceCount":1,"writtenSourceCount":1,"writeAttempts":0,"isSolo":true,"entitiesJson":"{\"anime_titles\":[],\"manga_titles\":[],\"work_titles\":[\"ArtiFixer\"],\"studios\":[\"NVIDIA\"],\"people\":[],\"type\":\"announcement\",\"domain\":\"other\",\"is_roundup\":false}","contentType":"news","status":"published","firstSeenAt":"2026-06-23T07:04:00.000Z","lastSeenAt":"2026-06-23T07:04:00.000Z","updatedAt":"2026-06-23T07:40:25.000Z"},"attribution":[{"source":"GIGAZINE","url":"https://gigazine.net/news/20260623-nvidia-artifixer-3d-representation/","title":"複数の写真から3Dシーンを生成するAIモデル「ArtiFixer」をNVIDIAが発表、写真にない部分も生成処理で補完可能"}],"entities":{"anime_titles":[],"manga_titles":[],"work_titles":["ArtiFixer"],"studios":["NVIDIA"],"people":[],"type":"announcement","domain":"other","is_roundup":false},"keyFacts":["Nvidia's ArtiFixer AI model has approximately 16.9 billion parameters.","ArtiFixer is based on the Wan 2.1 video generation AI.","The model was trained in two stages: first to generate and insert missing parts, then distilled into an autoregressive model.","ArtiFixer comes in three variants: ArtiFixer, ArtiFixer3D, and ArtiFixer3D+.","Nvidia says ArtiFixer3D+ outperforms conventional methods like 3DGUT, GenFusion, and GSFixer in quality."]}