Summary
InFlux++ is a computer vision resource for estimating per-frame camera intrinsics from video, addressing the common limitation that most 3D reconstruction algorithms assume fixed intrinsics while real-world footage often does not. The release consists of two components: InFlux++ Synth, a large-scale procedurally generated synthetic video dataset with over 441,000 annotated frames and 1,841 high-resolution videos providing accurate per-frame ground-truth intrinsics; and InFlux++ Real, an expanded real-world benchmark adding over 514,000 frames and 334 high-resolution videos. Fine-tuning existing intrinsic prediction methods on InFlux++ Synth improved focal length estimation accuracy on both real-world benchmarks, demonstrating that synthetic supervision is a promising approach for RGB-based camera intrinsics prediction. The paper is authored by Erich Liang, Caleb Kha-Uong, Chinmaya Saran, Sreemanti Dey, David W. Liu, Junhan Ouyang, Benjamin Zhou, and Jia Deng, and is available on arXiv (2607.05389).
Paper Overview
Field: Computer Vision
Authors: Erich Liang, Caleb Kha-Uong, Chinmaya Saran, Sreemanti Dey, David W. Liu, Junhan Ouyang, Benjamin Zhou, Jia Deng
Released: 2026-07-06
arXiv: 2607.05389
Abstract
Camera intrinsics are essential for recovering 3D structure from 2D video. Most 3D algorithms assume that video intrinsics are fixed, but real-world videos often violate this assumption.
InFlux++ consists of two parts:
- InFlux++ Synth — a large-scale procedurally generated synthetic video dataset containing 441,000+ annotated frames and 1,841 high-resolution videos, with accurate per-frame ground-truth intrinsics.
- InFlux++ Real — an expanded real-world benchmark adding 514,000+ frames and 334 high-resolution videos.
Fine-tuning existing intrinsics prediction methods on InFlux++ Synth improves focal length estimation accuracy on both real-world benchmarks, indicating that synthetic supervision is a promising direction for RGB-based camera intrinsics prediction.
Project website: https://influx.cs.princeton.edu/
This page is an English static mirror generated for search and AI citation.
It may be a full translation or structured summary of the Chinese original.
Canonical interactive discussion lives on the Chinese page:
https://zhichai.net/topic/178346201