GLOMAP – Global Structure-from-Motion Revisited
The paper introduces GLOMAP, a new system for 3D structure recovery and camera motion estimation, outperforming COLMAP in accuracy and speed, and is available as open-source software.
Read original article
The paper "Global Structure-from-Motion Revisited" by Linfei Pan and colleagues presents a new system called GLOMAP for recovering 3D structure and camera motion from images, a key area in computer vision known as Structure-from-Motion (SfM). Traditionally, SfM methods are divided into incremental and global approaches, with incremental methods being favored for their accuracy and robustness. However, global methods offer greater scalability and efficiency. GLOMAP aims to bridge this gap by providing a general-purpose system that not only matches but often surpasses the accuracy and robustness of COLMAP, the leading incremental SfM system, while being significantly faster. The GLOMAP pipeline consists of two main components: correspondence search and global estimation, with three innovative steps: view graph calibration, global positioning, and structure refinement. The core of GLOMAP is the Global Positioning step, where camera positions and image points are estimated from random positions. The system has been tested on the LaMAR dataset, which includes over 36,000 images. GLOMAP achieved a 90% recall rate at 1 meter within 5.5 hours, compared to COLMAP's approximately 50% recall rate and over 7 days of processing time. The paper emphasizes the efficiency and effectiveness of GLOMAP in the context of global SfM, making it a significant advancement in the field. The authors have made the system available as open-source software, contributing to ongoing research and development in computer vision.
Related
New computational microscopy technique provides direct route to crisp images
A team at the California Institute of Technology developed Angular Ptychographic Imaging with Closed-form method (APIC) for clearer microscopy images. APIC corrects aberrations efficiently, promising advancements in biomedical imaging and AI.
MASt3R – Matching and Stereo 3D Reconstruction
MASt3R, a model within the DUSt3R framework, excels in 3D reconstruction and feature mapping for image collections. It enhances depth perception, reduces errors, and revolutionizes spatial awareness across industries.
WildGaussians: 3D Gaussian Splatting in the Wild
A novel method, WildGaussians, enhances 3D scene reconstruction for in-the-wild data by combining DINO features and appearance modeling with 3D Gaussian Splatting. It outperforms NeRFs and 3DGS in handling dynamic scenes.
ColPali: Efficient Document Retrieval with Vision Language Models
The paper introduces ColPali, a model enhancing document retrieval by leveraging visual cues. It outperforms existing systems, introducing ViDoRe benchmark for evaluation, showcasing superior performance and speed in visually rich document retrieval.
Shape of Motion: 4D Reconstruction from a Single Video
Shape of Motion reconstructs 4D scenes from monocular videos by modeling 3D motion. It uses SE(3) motion bases and data-driven priors for global consistency. The method excels in 3D/2D motion estimation and view synthesis, as shown in experiments. Comparisons with other methods are included.
0 commentsRelated
New computational microscopy technique provides direct route to crisp images
A team at the California Institute of Technology developed Angular Ptychographic Imaging with Closed-form method (APIC) for clearer microscopy images. APIC corrects aberrations efficiently, promising advancements in biomedical imaging and AI.
MASt3R – Matching and Stereo 3D Reconstruction
MASt3R, a model within the DUSt3R framework, excels in 3D reconstruction and feature mapping for image collections. It enhances depth perception, reduces errors, and revolutionizes spatial awareness across industries.
WildGaussians: 3D Gaussian Splatting in the Wild
A novel method, WildGaussians, enhances 3D scene reconstruction for in-the-wild data by combining DINO features and appearance modeling with 3D Gaussian Splatting. It outperforms NeRFs and 3DGS in handling dynamic scenes.
ColPali: Efficient Document Retrieval with Vision Language Models
The paper introduces ColPali, a model enhancing document retrieval by leveraging visual cues. It outperforms existing systems, introducing ViDoRe benchmark for evaluation, showcasing superior performance and speed in visually rich document retrieval.
Shape of Motion: 4D Reconstruction from a Single Video
Shape of Motion reconstructs 4D scenes from monocular videos by modeling 3D motion. It uses SE(3) motion bases and data-driven priors for global consistency. The method excels in 3D/2D motion estimation and view synthesis, as shown in experiments. Comparisons with other methods are included.