利用场景语境进行大规模多人三维人体运动预测

Felix B Mueller, Julian Tanke, Juergen Gall
{"title":"利用场景语境进行大规模多人三维人体运动预测","authors":"Felix B Mueller, Julian Tanke, Juergen Gall","doi":"arxiv-2409.12189","DOIUrl":null,"url":null,"abstract":"Forecasting long-term 3D human motion is challenging: the stochasticity of\nhuman behavior makes it hard to generate realistic human motion from the input\nsequence alone. Information on the scene environment and the motion of nearby\npeople can greatly aid the generation process. We propose a scene-aware social\ntransformer model (SAST) to forecast long-term (10s) human motion motion.\nUnlike previous models, our approach can model interactions between both widely\nvarying numbers of people and objects in a scene. We combine a temporal\nconvolutional encoder-decoder architecture with a Transformer-based bottleneck\nthat allows us to efficiently combine motion and scene information. We model\nthe conditional motion distribution using denoising diffusion models. We\nbenchmark our approach on the Humans in Kitchens dataset, which contains 1 to\n16 persons and 29 to 50 objects that are visible simultaneously. Our model\noutperforms other approaches in terms of realism and diversity on different\nmetrics and in a user study. Code is available at\nhttps://github.com/felixbmuller/SAST.","PeriodicalId":501130,"journal":{"name":"arXiv - CS - Computer Vision and Pattern Recognition","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Massively Multi-Person 3D Human Motion Forecasting with Scene Context\",\"authors\":\"Felix B Mueller, Julian Tanke, Juergen Gall\",\"doi\":\"arxiv-2409.12189\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Forecasting long-term 3D human motion is challenging: the stochasticity of\\nhuman behavior makes it hard to generate realistic human motion from the input\\nsequence alone. Information on the scene environment and the motion of nearby\\npeople can greatly aid the generation process. We propose a scene-aware social\\ntransformer model (SAST) to forecast long-term (10s) human motion motion.\\nUnlike previous models, our approach can model interactions between both widely\\nvarying numbers of people and objects in a scene. We combine a temporal\\nconvolutional encoder-decoder architecture with a Transformer-based bottleneck\\nthat allows us to efficiently combine motion and scene information. We model\\nthe conditional motion distribution using denoising diffusion models. We\\nbenchmark our approach on the Humans in Kitchens dataset, which contains 1 to\\n16 persons and 29 to 50 objects that are visible simultaneously. Our model\\noutperforms other approaches in terms of realism and diversity on different\\nmetrics and in a user study. Code is available at\\nhttps://github.com/felixbmuller/SAST.\",\"PeriodicalId\":501130,\"journal\":{\"name\":\"arXiv - CS - Computer Vision and Pattern Recognition\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-18\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Computer Vision and Pattern Recognition\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.12189\",\"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 - Computer Vision and Pattern Recognition","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.12189","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

预测长期三维人体运动具有挑战性:由于人体行为具有随机性,因此很难仅凭输入序列生成逼真的人体运动。有关场景环境和附近人员运动的信息可以极大地帮助生成过程。与之前的模型不同,我们的方法可以模拟场景中数量变化很大的人和物体之间的互动。我们将时空卷积编码器-解码器架构与基于变换器的瓶颈相结合,从而有效地结合了运动和场景信息。我们使用去噪扩散模型对条件运动分布进行建模。我们在 "厨房中的人类 "数据集上对我们的方法进行了测试,该数据集包含 1 到 16 个人和 29 到 50 个同时可见的物体。我们的模型在不同指标的逼真度和多样性方面,以及在用户研究中,都优于其他方法。代码可在https://github.com/felixbmuller/SAST。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Massively Multi-Person 3D Human Motion Forecasting with Scene Context
Forecasting long-term 3D human motion is challenging: the stochasticity of human behavior makes it hard to generate realistic human motion from the input sequence alone. Information on the scene environment and the motion of nearby people can greatly aid the generation process. We propose a scene-aware social transformer model (SAST) to forecast long-term (10s) human motion motion. Unlike previous models, our approach can model interactions between both widely varying numbers of people and objects in a scene. We combine a temporal convolutional encoder-decoder architecture with a Transformer-based bottleneck that allows us to efficiently combine motion and scene information. We model the conditional motion distribution using denoising diffusion models. We benchmark our approach on the Humans in Kitchens dataset, which contains 1 to 16 persons and 29 to 50 objects that are visible simultaneously. Our model outperforms other approaches in terms of realism and diversity on different metrics and in a user study. Code is available at https://github.com/felixbmuller/SAST.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信