MASt3R
PulseAugur coverage of MASt3R — every cluster mentioning MASt3R across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
MuViSeg advances multi-view segment matching for improved navigation
Researchers have developed MuViSeg, a novel approach for matching segments across multiple image views, improving upon existing methods that rely on pairwise comparisons. The system incorporates learned matching heads, …
-
New RASR method enhances UAV navigation without GNSS
Researchers have developed a new method called Range-Aware Scale Recovery (RASR) to improve metric navigation for unmanned aerial vehicles (UAVs) when Global Navigation Satellite System (GNSS) signals are unavailable. R…
-
MAC-Splat framework enhances 3D reconstruction from sparse views
Researchers have introduced MAC-Splat, a novel training framework designed to improve the fidelity of 3D scene reconstruction from sparse camera views. This method addresses limitations in existing 3D Gaussian Splatting…
-
RayTun3R adapts 3D foundation models for fisheye cameras
Researchers have developed RayTun3R, a novel method to adapt existing 3D foundation models for use with fisheye camera imagery. These models, which typically perform well with standard pinhole cameras, degrade significa…
-
AI model distillation enables efficient 3D reconstruction for space exploration
Researchers have developed a method to distill large 3D foundation models, like MASt3R, into smaller, more efficient versions for applications with limited computing power, such as lunar exploration. By fine-tuning a MA…
-
New Moonstone Benchmark and Model Advance Lunar Remote Sensing
Researchers have developed Moonstone, a multimodal foundation model and benchmark specifically designed for lunar remote sensing. This initiative addresses the fragmentation of lunar datasets and the lack of standardize…
-
VGGT Model Uncertainty Quality Analyzed for Improved 3D Reconstruction
A new paper analyzes the uncertainty quality of the Visual Geometry Grounded Transformer (VGGT) model, which recently won a Best Paper Award at CVPR 2025. The research identifies a confidence threshold for filtering VGG…
-
New SADGE metric predicts synthetic data performance in computer vision
Researchers have developed SADGE, a new metric designed to predict how well synthetic image datasets will perform on real-world computer vision tasks. Unlike previous methods that focused on either appearance or geometr…
-
New benchmark evaluates 3D reconstruction consistency amid AI hallucinations
Researchers have developed a new benchmark, \benchmark, to evaluate the consistency of 3D reconstructions from multiple camera views, particularly when 3D foundation models hallucinate details. This benchmark compares n…
-
WildPose framework enhances pose estimation in dynamic environments
Researchers have introduced WildPose, a novel monocular pose estimation framework designed to operate effectively in dynamic environments. This unified approach combines the perceptual capabilities of feedforward models…
-
Researchers test pretrained image matchers for satellite registration tasks
Researchers investigated the effectiveness of twenty-four pretrained image matching models for cross-modal SAR-optical satellite registration, a crucial step for remote sensing in disaster response. Their findings indic…
-
REALM framework aligns RGB and event camera data for cross-modal perception
Researchers have developed REALM, a novel cross-modal framework designed to align RGB and event camera data within a shared latent manifold. This approach projects event representations into the latent space of pre-trai…