Adaptive Annealing for Robust Averaging

Chitturi, S and Govindu, VM (2025) Adaptive Annealing for Robust Averaging. In: 18th European Conference on Computer Vision, ECCV 2024, 29 September 2024through 4 October 2024, Milan, pp. 53-69.

PDF lec_not_com_sci_15128_2025.pdf - Published Version Restricted to Registered users only Download (701kB) | Request a copy

Abstract

Graduated Non-Convexity (GNC) is a robust estimation method in which an objective function is progressively annealed starting from a smooth convex form to one that represents the desired objective function. Such annealing is achieved by modifying a scale parameter in the objective function that is solved at each stage. A fixed annealing scheme often leads to a poor efficiency vs accuracy tradeoff, whereas adaptive annealing lacks scalability for large scale problems. An important large scale estimation problem is that of averaging that arises in the context of 3D reconstruction, wherein pairwise observations on edges of a viewgraph are used to estimate the corresponding values at the graph vertices. In this paper we present a novel adaptive GNC framework tailored for averaging problems in vector spaces, specifically vector and translation averaging. Our approach leverages insights from graph Laplacians and imparts scalability. We demonstrate the superior performance of our adaptive approach while maintaining efficiency in comparison to baselines. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Item Type:	Conference Paper
Publication:	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher:	Springer Science and Business Media Deutschland GmbH
Additional Information:	The copyright for this article belongs to publishers.
Keywords:	Graph theory; Laplace transforms; Scalability; Simulated annealing, Averaging; Graduated non convexities; Graduated non-convexity; Graph Laplacian; Graph laplacians; Large-scale problem; Objective functions; Poor efficiencies; Robust estimation method; Scale parameter, Vector spaces
Department/Centre:	Division of Electrical Sciences > Electrical Engineering
Date Deposited:	30 Dec 2024 06:12
Last Modified:	30 Dec 2024 06:12
URI:	http://eprints.iisc.ac.in/id/eprint/87178

Actions (login required)

View Item