Abstract:In this paper, we present DM-Calib, a diffusion-based approach for estimating pin- hole camera intrinsic parameters from a single input image. Monocular camera calibration is essential for many 3D vision tasks. However, most existing methods depend on handcrafted assumptions or are constrained by limited training data, re- sulting in poor generalization across diverse real-world images. Recent advance- ments in stable diffusion models, trained on massive data, have shown the ability to generate high-quality images with varied characteristics. Emerging evidence in- dicates that these models implicitly capture the relationship between camera focal length and image content. Building on this insight, we explore how to leverage the powerful priors of diffusion models for monocular pinhole camera calibration. Specifically, we introduce a new image-based representation, termed Camera Im- age, which losslessly encodes the numerical camera intrinsics and integrates seam- lessly with the diffusion framework. Using this representation, we reformulate the problem of estimating camera intrinsics as the generation of a dense Camera Im- age conditioned on an input image. By fine-tuning a stable diffusion model to gen- erate a Camera Image from a single RGB input, we can extract camera intrinsics via a RANSAC operation. We further demonstrate that our monocular calibration method enhances performance across various 3D tasks, including zero-shot metric depth estimation, 3D metrology, pose estimation and sparse-view reconstruction. Extensive experiments on multiple public datasets show that our approach signifi- cantly outperforms baselines and provides broad benefits to 3D vision tasks. Code is available at <a class="link-external link-https" href="https://github.com/JunyuanDeng/DM-Calib" rel="external noopener nofollow">this https URL</a>.

ON THE CAMERA POSITION DITHERING IN VISUAL 3D RECONSTRUCTION

Multi-view 3D Reconstruction from Uncalibrated Radially-Symmetric Cameras

A Simple 3D Reconstruction Method Based on Single Camera

Camera Configuration Design in Cooperative Active Visual 3d Reconstruction: A Statistical Approach

Asymmetric representation for 3D panoramic video

The analysis of the structural parameter influences on measurement errors in a binocular 3D reconstruction system: a portable 3D system

Robust 3D Reconstruction with an RGB-D Camera

Three-Dimensional Reconstruction Based on Quadrotors with Binocular Vision

Real-time High-accuracy Three-Dimensional Reconstruction with Consumer RGB-D Cameras

Boost 3D Reconstruction using Diffusion-based Monocular Camera Calibration

Real-Time 3D Reconstruction Method Based on Monocular Vision

Real-time Indoor Scene Reconstruction with RGBD and Inertial Input.

A binocular stereo-view 3D reconstruction algorithm based on contour extraction and depth screening

Impact of Camera Settings on 3D Reconstruction Quality: Insights from NeRF and Gaussian Splatting

Improving the Noise Immunity of 3D Computational Ghost Imaging.

Volumetric Calculation of Quantization Error in 3-D Vision Systems

Binocular Rivalry - Psychovisual Challenge in Stereoscopic Video Error Concealment

Noise-Resilient Reconstruction of Panoramas and 3D Scenes Using Robot-Mounted Unsynchronized Commodity RGB-D Cameras

Omnidirectional-Sensor-System-Based Texture Noise Correction in Large-Scale 3D Reconstruction

On 3D Reconstruction Using RGB-D Cameras

Real-Time 3D Eye Performance Reconstruction for RGBD Cameras