[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84830-en":3,"doc-seo-84830-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},84830,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",8,"Research & Report","InFlux++ Real and Synthetic Data for Estimating Dynamic Camera Intrinsics","Camera intrinsics are essential for recovering 3D structure from 2D video, yet most 3D methods assume intrinsics stay fixed, which often breaks for in-the-wild dynamic footage. InFlux++ targets accurate per-frame intrinsics estimation from RGB. It builds two datasets: InFlux++ Synth, a procedurally generated set with 441K+ annotated frames and per-frame ground-truth intrinsics (plus pose, depth, and normals for a subset), and InFlux++ Real, a real-world benchmark with 514K+ frames across broader scenes and camera motions. Fine-tuning improves focal length estimation.","arXiv :2607 .05389v 1 [ cs .CV] 6 Jul 2026  \nInFlux++: Real and Synthetic Data for Estimating Dynamic Camera Intrinsics  \nErich Liang, Caleb Kha-Uong⋆, Chinmaya Saran⋆, Sreemanti Dey⋆, David W. Liu, Junhan Ouyang, Benjamin Zhou, and Jia Deng  \nPrinceton University, Princeton NJ 08544, USA {erliang,ck2867,cs4046,sd9968,dl3533, harryouyang,bz2883,[jiadeng}@princeton.edu](jiadeng}@princeton.edu)  \nAbstract. Camera intrinsics play a vital role in recovering 3D structure from 2D video. However, most 3D algorithms assume that intrinsics remain fixed throughout input video, an assumption that frequently fails for real-world in-the-wild videos. Consequently, estimating per-frame intrinsics from RGB images is critical for enabling 3D methods to operate robustly on dynamic intrinsics videos. Previously, InFlux has contributed to this research direction by establishing the first real-world benchmark with per-frame ground truth intrinsics for dynamic intrinsics videos, enabling systematic evaluation of algorithms for this task. Nevertheless, existing methods remain inaccurate due to two key obstacles: (i) training data for this task is scarce and lacks diversity in camera intrinsics;  \nand (ii) existing benchmarks, including InFlux, are limited in scene and camera motion diversity, making it difficult to properly evaluate method performance. To address both gaps, we present InFlux++, consisting of two components. InFlux++ Synth is a large-scale procedurally generated synthetic video dataset with 441K+ annotated frames from 1841 highresolution videos, providing accurate per-frame ground truth intrinsics for training dynamic intrinsics prediction models; a subset additionally includes per-frame camera pose, depth, and surface normals. The videos feature rich intrinsics diversity through changes in camera zoom and focus over time, as well as dynamic objects and realistic rendering effects such as lens distortion and defocus blur. InFlux++ Real is a largescale real-world benchmark that extends InFlux with 514K+ newly captured frames across 334 high-resolution videos, spanning a wider range of scenes and camera motions. Finetuning existing intrinsics prediction methods on InFlux++ Synth consistently improves focal length estimation across both InFlux++ Real and InFlux, suggesting that synthetic supervision is a promising direction for RGB-based intrinsics prediction.  \nFor the dataset, benchmark, code, videos, submission instructions, and live leaderboard, please visit [https://influx.cs.princeton.edu/](https://influx.cs.princeton.edu/) .  \nKeywords: Dynamic Camera Intrinsics · Real-world and Synthetic Video  \n· Procedural Generation  \n⋆ Equal contribution.  \n2 E. Liang et al.  \nInFlux++ Synth  \nInFlux++ Real  \nFig. 1: A gallery of InFlux++, our data suite for dynamic intrinsics prediction. The top three rows show videos from InFlux++ Synth, a synthetic dataset with 441K+ frames across 1841 videos, featuring diverse intrinsics from changing zoom and focus, varying lens distortion, dynamic objects, and realistic optical effects such as defocus blur. The bottom three rows show videos from InFlux++ Real, a real-world dynamic intrinsics benchmark with 514K+ frames across 334 high-resolution videos. Compared to InFlux [12], it captures a wider range of indoor and outdoor scenes, natural daily activities, and richer camera motions with more translational motion.  \n1 Introduction  \nCamera intrinsics are fundamental to many real-world 3D systems, as they define the geometric mapping between 3D coordinates and the 2D image. Example applications include robotic depth perception and digital overlays in AR/VR.  \nMany modern 3D algorithms [8,14,15,23,26,27] assume that camera intrinsics remain constant throughout video input, but real-world in-the-wild footage often violates this assumption. Intrinsics can vary during capture due to changes in zoom or focus distance, which alter the optical configuration of the lens.  \nAs a result, accurately predicting pe","cbCaihyhgTo5Jfly","https://ap.wps.com/l/cbCaihyhgTo5Jfly","pdf",30958170,1,37,"English","en",105,"# Introduction\n## Motivation: dynamic intrinsics in the wild\n## InFlux and limitations\n## InFlux++ overview (Synth and Real)\n## Dataset details and evaluation direction","[{\"question\":\"Why is estimating per-frame camera intrinsics important for dynamic intrinsics video?\",\"answer\":\"Most 3D algorithms assume fixed intrinsics, but real videos often change intrinsics during capture via zoom or focus. Per-frame estimation enables downstream geometric computations to use correct calibration values and improves reliability.\"},{\"question\":\"What is included in InFlux++ Synth?\",\"answer\":\"InFlux++ Synth is a procedurally generated synthetic dataset with 441K+ annotated frames from 1841 high-resolution videos, providing accurate per-frame ground-truth intrinsics. A subset also includes per-frame camera pose, depth, and surface normals.\"},{\"question\":\"How does InFlux++ Real expand the benchmarking scope?\",\"answer\":\"InFlux++ Real adds 514K+ newly captured frames across 334 high-resolution videos, covering a wider range of scenes and camera motions, including richer translational motion compared with InFlux.\"}]",1784198590,93,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"influx-real-and-synthetic-data-for-estimating-dynamic-camera-intrinsics","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/influx-real-and-synthetic-data-for-estimating-dynamic-camera-intrinsics/84830/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is estimating per-frame camera intrinsics important for dynamic intrinsics video?","Question",{"text":75,"@type":76},"Most 3D algorithms assume fixed intrinsics, but real videos often change intrinsics during capture via zoom or focus. Per-frame estimation enables downstream geometric computations to use correct calibration values and improves reliability.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is included in InFlux++ Synth?",{"text":80,"@type":76},"InFlux++ Synth is a procedurally generated synthetic dataset with 441K+ annotated frames from 1841 high-resolution videos, providing accurate per-frame ground-truth intrinsics. A subset also includes per-frame camera pose, depth, and surface normals.",{"name":82,"@type":73,"acceptedAnswer":83},"How does InFlux++ Real expand the benchmarking scope?",{"text":84,"@type":76},"InFlux++ Real adds 514K+ newly captured frames across 334 high-resolution videos, covering a wider range of scenes and camera motions, including richer translational motion compared with InFlux.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]