Sign up to save your podcasts
Or
Share $Ï€^3$: Scalable Permutation-Equivariant Visual Geometry Learning
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
$Ï€^3$: Scalable Permutation-Equivariant Visual Geometry Learning
🤗 Upvotes: 36 | cs.CV
Authors:
Yifan Wang, Jianjun Zhou, Haoyi Zhu, Wenzheng Chang, Yang Zhou, Zizun Li, Junyi Chen, Jiangmiao Pang, Chunhua Shen, Tong He
Title:
$Ï€^3$: Scalable Permutation-Equivariant Visual Geometry Learning
Arxiv:
http://arxiv.org/abs/2507.13347v1
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.
View all episodes
By Jingwen Liang, Gengyu Wang🤗 Upvotes: 36 | cs.CV
Authors:
Yifan Wang, Jianjun Zhou, Haoyi Zhu, Wenzheng Chang, Yang Zhou, Zizun Li, Junyi Chen, Jiangmiao Pang, Chunhua Shen, Tong He
Title:
$Ï€^3$: Scalable Permutation-Equivariant Visual Geometry Learning
Arxiv:
http://arxiv.org/abs/2507.13347v1
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.