{"title":"基于神经过程的连续条件视频合成","authors":"Xi Ye, Guillaume-Alexandre Bilodeau","doi":"10.1016/j.cviu.2025.104387","DOIUrl":null,"url":null,"abstract":"<div><div>Different conditional video synthesis tasks, such as frame interpolation and future frame prediction, are typically addressed individually by task-specific models, despite their shared underlying characteristics. Additionally, most conditional video synthesis models are limited to discrete frame generation at specific integer time steps. This paper presents a unified model that tackles both challenges simultaneously. We demonstrate that conditional video synthesis can be formulated as a neural process, where input spatio-temporal coordinates are mapped to target pixel values by conditioning on context spatio-temporal coordinates and pixel values. Our approach leverages a Transformer-based non-autoregressive conditional video synthesis model that takes the implicit neural representation of coordinates and context pixel features as input. Our task-specific models outperform previous methods for future frame prediction and frame interpolation across multiple datasets. Importantly, our model enables temporal continuous video synthesis at arbitrary high frame rates, outperforming the previous state-of-the-art. The source code and video demos for our model are available at <span><span>https://npvp.github.io</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50633,"journal":{"name":"Computer Vision and Image Understanding","volume":"259 ","pages":"Article 104387"},"PeriodicalIF":4.3000,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Continuous conditional video synthesis by neural processes\",\"authors\":\"Xi Ye, Guillaume-Alexandre Bilodeau\",\"doi\":\"10.1016/j.cviu.2025.104387\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Different conditional video synthesis tasks, such as frame interpolation and future frame prediction, are typically addressed individually by task-specific models, despite their shared underlying characteristics. Additionally, most conditional video synthesis models are limited to discrete frame generation at specific integer time steps. This paper presents a unified model that tackles both challenges simultaneously. We demonstrate that conditional video synthesis can be formulated as a neural process, where input spatio-temporal coordinates are mapped to target pixel values by conditioning on context spatio-temporal coordinates and pixel values. Our approach leverages a Transformer-based non-autoregressive conditional video synthesis model that takes the implicit neural representation of coordinates and context pixel features as input. Our task-specific models outperform previous methods for future frame prediction and frame interpolation across multiple datasets. Importantly, our model enables temporal continuous video synthesis at arbitrary high frame rates, outperforming the previous state-of-the-art. The source code and video demos for our model are available at <span><span>https://npvp.github.io</span><svg><path></path></svg></span>.</div></div>\",\"PeriodicalId\":50633,\"journal\":{\"name\":\"Computer Vision and Image Understanding\",\"volume\":\"259 \",\"pages\":\"Article 104387\"},\"PeriodicalIF\":4.3000,\"publicationDate\":\"2025-05-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Computer Vision and Image Understanding\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1077314225001109\",\"RegionNum\":3,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computer Vision and Image Understanding","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1077314225001109","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
Continuous conditional video synthesis by neural processes
Different conditional video synthesis tasks, such as frame interpolation and future frame prediction, are typically addressed individually by task-specific models, despite their shared underlying characteristics. Additionally, most conditional video synthesis models are limited to discrete frame generation at specific integer time steps. This paper presents a unified model that tackles both challenges simultaneously. We demonstrate that conditional video synthesis can be formulated as a neural process, where input spatio-temporal coordinates are mapped to target pixel values by conditioning on context spatio-temporal coordinates and pixel values. Our approach leverages a Transformer-based non-autoregressive conditional video synthesis model that takes the implicit neural representation of coordinates and context pixel features as input. Our task-specific models outperform previous methods for future frame prediction and frame interpolation across multiple datasets. Importantly, our model enables temporal continuous video synthesis at arbitrary high frame rates, outperforming the previous state-of-the-art. The source code and video demos for our model are available at https://npvp.github.io.
期刊介绍:
The central focus of this journal is the computer analysis of pictorial information. Computer Vision and Image Understanding publishes papers covering all aspects of image analysis from the low-level, iconic processes of early vision to the high-level, symbolic processes of recognition and interpretation. A wide range of topics in the image understanding area is covered, including papers offering insights that differ from predominant views.
Research Areas Include:
• Theory
• Early vision
• Data structures and representations
• Shape
• Range
• Motion
• Matching and recognition
• Architecture and languages
• Vision systems