no code implementations • 1 Oct 2024 • Liangzu Peng, Juan Elenter, Joshua Agterberg, Alejandro Ribeiro, René Vidal
This results in a stable continual learning method with strong empirical performance and theoretical guarantees.
1 code implementation • 9 Apr 2024 • Tianyu Huang, Haoang Li, Liangzu Peng, Yinlong Liu, Yun-hui Liu
Our strategy largely reduces the search space and can guarantee accuracy with only a few inlier samples, therefore enjoying an excellent trade-off between efficiency and robustness.
1 code implementation • CVPR 2024 • Tianyu Huang, Liangzu Peng, René Vidal, Yun-hui Liu
Given an input set of $3$D point pairs, the goal of outlier-robust $3$D registration is to compute some rotation and translation that align as many point pairs as possible.
1 code implementation • 29 Apr 2023 • Liangzu Peng, Paris V. Giampouras, René Vidal
We show that ICL unifies multiple well-established continual learning methods and gives new theoretical insights into the strengths and weaknesses of these methods.
no code implementations • 11 Apr 2023 • Xinyue Zhang, Liangzu Peng, Wanting Xu, Laurent Kneip
Branch-and-bound-based consensus maximization stands out due to its important ability of retrieving the globally optimal solution to outlier-affected geometric problems.
1 code implementation • CVPR 2023 • Liangzu Peng, Christian Kümmerle, René Vidal
Outlier-robust estimation involves estimating some parameters (e. g., 3D rotations) from data samples in the presence of outliers, and is typically formulated as a non-convex and non-smooth problem.
no code implementations • 18 Jul 2022 • Liangzu Peng, Mahyar Fazlyab, René Vidal
To induce robustness against outliers for rotation search, prior work considers truncated least-squares (TLS), which is a non-convex optimization problem, and its semidefinite relaxation (SDR) as a tractable alternative.
1 code implementation • CVPR 2022 • Liangzu Peng, Manolis C. Tsakiris, René Vidal
We first propose a solver, $\texttt{ARCS}$, that i) assumes noiseless point sets in general position, ii) requires only $2$ inliers, iii) uses $O(m\log m)$ time and $O(m)$ space, and iv) can successfully solve the problem even with, e. g., $m, n\approx 10^6$ in about $0. 1$ seconds.
1 code implementation • NeurIPS 2021 • Yunzhen Yao, Liangzu Peng, Manolis C. Tsakiris
Allowing for missing entries on top of permutations in UPCA leads to the problem of unlabeled matrix completion, for which we derive theory and algorithms of similar flavor.
no code implementations • 9 Jun 2020 • Liangzu Peng, Manolis C. Tsakiris
In this paper, we provide tighter and simpler conditions that guarantee the unique recovery for the single-subspace case, extend the result to the case of a subspace arrangement, and show that the unique recovery in a single subspace is locally stable under noise.
1 code implementation • 17 Mar 2020 • Liangzu Peng, Manolis C. Tsakiris
Linear regression without correspondences concerns the recovery of a signal in the linear regression setting, where the correspondences between the observations and the linear functionals are unknown.
no code implementations • 12 Oct 2018 • Manolis C. Tsakiris, Liangzu Peng, Aldo Conca, Laurent Kneip, Yuanming Shi, Hayoung Choi
This naturally leads to a polynomial system of $n$ equations in $n$ unknowns, which contains $\xi^*$ in its root locus.