Related papers: $π^3$: Scalable Permutation-Equivariant Visual Geometry Learning

$π^3$: Scalable Permutation-Equivariant Visual Geometry Learning

URL: http://arxiv.org/abs/2507.13347v1
Date: Thu, 17 Jul 2025 17:59:53 GMT
Title: $π^3$: Scalable Permutation-Equivariant Visual Geometry Learning
Authors: Yifan Wang, Jianjun Zhou, Haoyi Zhu, Wenzheng Chang, Yang Zhou, Zizun Li, Junyi Chen, Jiangmiao Pang, Chunhua Shen, Tong He,
Abstract summary: $pi3$ is a feed-forward neural network that offers a novel approach to visual geometry reconstruction.<n>pi3$ employs a fully permutation-equivariant architecture to predict affine-invariant camera poses and scale-invariant local point maps.
Score: 50.80418813055225
License: http://creativecommons.org/licenses/by/4.0/
Abstract: We introduce $\pi^3$, a feed-forward neural network that offers a novel approach to visual geometry reconstruction, breaking the reliance on a conventional fixed reference view. Previous methods often anchor their reconstructions to a designated viewpoint, an inductive bias that can lead to instability and failures if the reference is suboptimal. In contrast, $\pi^3$ employs a fully permutation-equivariant architecture to predict affine-invariant camera poses and scale-invariant local point maps without any reference frames. This design makes our model inherently robust to input ordering and highly scalable. These advantages enable our simple and bias-free approach to achieve state-of-the-art performance on a wide range of tasks, including camera pose estimation, monocular/video depth estimation, and dense point map reconstruction. Code and models are publicly available.

Related papers

Dens3R: A Foundation Model for 3D Geometry Prediction [44.13431776180547]
Dens3R is a 3D foundation model designed for joint geometric dense prediction.<n>By integrating image-pair matching features with intrinsic invariance modeling, Dens3R accurately regresses multiple geometric quantities.
arXiv Detail & Related papers (2025-07-22T07:22:30Z)
A Guide to Structureless Visual Localization [63.41481414949785]
Methods that estimate the camera pose of a query image in a known scene are core components of many applications, including self-driving cars and augmented / mixed reality systems.<n>State-of-the-art visual localization algorithms are structure-based, i.e., they store a 3D model of the scene and use 2D-3D correspondences between the query image and 3D points in the model for camera pose estimation.<n>This paper is dedicated to providing, to the best of our knowledge, first comprehensive discussion and comparison of structureless methods.
arXiv Detail & Related papers (2025-04-24T15:08:36Z)
UNOPose: Unseen Object Pose Estimation with an Unposed RGB-D Reference Image [86.7128543480229]
Unseen object pose estimation methods often rely on CAD models or multiple reference views.<n>To simplify reference acquisition, we aim to estimate the unseen object's pose through a single unposed RGB-D reference image.<n>We present a novel approach and benchmark, termed UNOPose, for unseen one-reference-based object pose estimation.
arXiv Detail & Related papers (2024-11-25T05:36:00Z)
No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images [100.80376573969045]
NoPoSplat is a feed-forward model capable of reconstructing 3D scenes parameterized by 3D Gaussians from multi-view images. Our model achieves real-time 3D Gaussian reconstruction during inference. This work makes significant advances in pose-free generalizable 3D reconstruction and demonstrates its applicability to real-world scenarios.
arXiv Detail & Related papers (2024-10-31T17:58:22Z)
GenS: Generalizable Neural Surface Reconstruction from Multi-View Images [20.184657468900852]
GenS is an end-to-end generalizable neural surface reconstruction model. Our representation is more powerful, which can recover high-frequency details while maintaining global smoothness. Experiments on popular benchmarks show that our model can generalize well to new scenes.
arXiv Detail & Related papers (2024-06-04T17:13:10Z)
FrozenRecon: Pose-free 3D Scene Reconstruction with Frozen Depth Models [67.96827539201071]
We propose a novel test-time optimization approach for 3D scene reconstruction. Our method achieves state-of-the-art cross-dataset reconstruction on five zero-shot testing datasets.
arXiv Detail & Related papers (2023-08-10T17:55:02Z)
Equivariant Light Field Convolution and Transformer [40.840098156362316]
Deep learning of geometric priors from 2D images often requires each image to be represented in a $2D$ canonical frame. We show how to learn priors from multiple views equivariant to coordinate frame transformations by proposing an $SE(3)$-equivariant convolution and transformer in the space of rays in 3D.
arXiv Detail & Related papers (2022-12-30T18:38:31Z)
Pixel-Perfect Structure-from-Motion with Featuremetric Refinement [96.73365545609191]
We refine two key steps of structure-from-motion by a direct alignment of low-level image information from multiple views. This significantly improves the accuracy of camera poses and scene geometry for a wide range of keypoint detectors. Our system easily scales to large image collections, enabling pixel-perfect crowd-sourced localization at scale.
arXiv Detail & Related papers (2021-08-18T17:58:55Z)
A Divide et Impera Approach for 3D Shape Reconstruction from Multiple Views [49.03830902235915]
Estimating the 3D shape of an object from a single or multiple images has gained popularity thanks to the recent breakthroughs powered by deep learning. This paper proposes to rely on viewpoint variant reconstructions by merging the visible information from the given views. To validate the proposed method, we perform a comprehensive evaluation on the ShapeNet reference benchmark in terms of relative pose estimation and 3D shape reconstruction.
arXiv Detail & Related papers (2020-11-17T09:59:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.