MIXER: A Principled Framework for Multimodal, Multiway Data Association
arXiv:2111.14990
Abstract
A fundamental problem in robotic perception is matching identical objects or data, with applications such as loop closure detection, place recognition, object tracking, and map fusion. While the problem becomes considerably more challenging when matching should be done jointly across multiple, multimodal sets of data, the robustness and accuracy of matching in the presence of noise and outliers can be greatly improved in this setting. At present, multimodal techniques do not leverage multiway information, and multiway techniques do not incorporate different modalities, leading to inferior results. In contrast, we present a principled mixed-integer quadratic framework to address this issue. We use a novel continuous relaxation in a projected gradient descent algorithm that guarantees feasible solutions of the integer program are obtained efficiently. We demonstrate experimentally that correspondences obtained from our approach are more stable to noise and errors than state-of-the-art techniques. Tested on a robotics dataset, our algorithm resulted in a 35% increase in F1 score when compared to the best alternative.
arXiv commentpresented in ICRA 2021 Workshop on Robust Perception for Autonomous Field Robots in Challenging Environments
References in corpus (6)
- YOLOv3: An Incremental Improvement
- MOT16: A Benchmark for Multi-Object Tracking
- Multiple Object Tracking: A Literature Review
- A General Multi-Graph Matching Approach via Graduated Consistency-regularized Boosting
- Learn to Combine Modalities in Multimodal Deep Learning
- Synchronisation of Partial Multi-Matchings via Non-negative Factorisations