Yiwen Chen, Yikai Wang, Yihao Luo, Zhengyi Wang, Zilong Chen, Jun Zhu, Chi Zhang, Guosheng Lin
{"title":"MeshAnything V2:通过相邻网格标记化生成艺术家创作的网格","authors":"Yiwen Chen, Yikai Wang, Yihao Luo, Zhengyi Wang, Zilong Chen, Jun Zhu, Chi Zhang, Guosheng Lin","doi":"arxiv-2408.02555","DOIUrl":null,"url":null,"abstract":"We introduce MeshAnything V2, an autoregressive transformer that generates\nArtist-Created Meshes (AM) aligned to given shapes. It can be integrated with\nvarious 3D asset production pipelines to achieve high-quality, highly\ncontrollable AM generation. MeshAnything V2 surpasses previous methods in both\nefficiency and performance using models of the same size. These improvements\nare due to our newly proposed mesh tokenization method: Adjacent Mesh\nTokenization (AMT). Different from previous methods that represent each face\nwith three vertices, AMT uses a single vertex whenever possible. Compared to\nprevious methods, AMT requires about half the token sequence length to\nrepresent the same mesh in average. Furthermore, the token sequences from AMT\nare more compact and well-structured, fundamentally benefiting AM generation.\nOur extensive experiments show that AMT significantly improves the efficiency\nand performance of AM generation. Project Page:\nhttps://buaacyw.github.io/meshanything-v2/","PeriodicalId":501174,"journal":{"name":"arXiv - CS - Graphics","volume":"27 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"MeshAnything V2: Artist-Created Mesh Generation With Adjacent Mesh Tokenization\",\"authors\":\"Yiwen Chen, Yikai Wang, Yihao Luo, Zhengyi Wang, Zilong Chen, Jun Zhu, Chi Zhang, Guosheng Lin\",\"doi\":\"arxiv-2408.02555\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We introduce MeshAnything V2, an autoregressive transformer that generates\\nArtist-Created Meshes (AM) aligned to given shapes. It can be integrated with\\nvarious 3D asset production pipelines to achieve high-quality, highly\\ncontrollable AM generation. MeshAnything V2 surpasses previous methods in both\\nefficiency and performance using models of the same size. These improvements\\nare due to our newly proposed mesh tokenization method: Adjacent Mesh\\nTokenization (AMT). Different from previous methods that represent each face\\nwith three vertices, AMT uses a single vertex whenever possible. Compared to\\nprevious methods, AMT requires about half the token sequence length to\\nrepresent the same mesh in average. Furthermore, the token sequences from AMT\\nare more compact and well-structured, fundamentally benefiting AM generation.\\nOur extensive experiments show that AMT significantly improves the efficiency\\nand performance of AM generation. Project Page:\\nhttps://buaacyw.github.io/meshanything-v2/\",\"PeriodicalId\":501174,\"journal\":{\"name\":\"arXiv - CS - Graphics\",\"volume\":\"27 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-08-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Graphics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2408.02555\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Graphics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2408.02555","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
我们介绍 MeshAnything V2,它是一种自回归变换器,可生成与给定形状对齐的艺术家自创网格(AM)。它可以与各种三维资产生产流水线集成,以实现高质量、高度可控的 AM 生成。在使用相同大小的模型时,MeshAnything V2 在效率和性能上都超越了以前的方法。这些改进归功于我们新提出的网格标记化方法:相邻网格标记化(AMT)。与以前用三个顶点表示每个面的方法不同,AMT 尽可能使用单个顶点。与以前的方法相比,AMT 平均只需要一半的标记序列长度就能表示相同的网格。我们的大量实验表明,AMT 显著提高了 AM 生成的效率和性能。项目页面:https://buaacyw.github.io/meshanything-v2/
MeshAnything V2: Artist-Created Mesh Generation With Adjacent Mesh Tokenization
We introduce MeshAnything V2, an autoregressive transformer that generates
Artist-Created Meshes (AM) aligned to given shapes. It can be integrated with
various 3D asset production pipelines to achieve high-quality, highly
controllable AM generation. MeshAnything V2 surpasses previous methods in both
efficiency and performance using models of the same size. These improvements
are due to our newly proposed mesh tokenization method: Adjacent Mesh
Tokenization (AMT). Different from previous methods that represent each face
with three vertices, AMT uses a single vertex whenever possible. Compared to
previous methods, AMT requires about half the token sequence length to
represent the same mesh in average. Furthermore, the token sequences from AMT
are more compact and well-structured, fundamentally benefiting AM generation.
Our extensive experiments show that AMT significantly improves the efficiency
and performance of AM generation. Project Page:
https://buaacyw.github.io/meshanything-v2/