Q-Match: Iterative Shape Matching via Quantum Annealing

Seelbach, Benkner, Marcel; Lähner, Zorah; Golyanik, Vladislav; Wunderlich, Christof; Theobalt, Christian; Moeller, Michael

doi:10.1109/iccv48922.2021.00749

Cited by 20 publications

(28 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Benkner et al [3] use adiabatic quantum computing to match 3D shapes and images with permutation matrices and investigate different constraint formulations to optimize the probability of finding a correct solution. By using an iterative approach, the same authors are able to scale the approach up to larger problem instances [4]. Closest to our work is the contribution of Birdal et al [6].…”

Section: Related Workmentioning

confidence: 77%

“…While widely used flow formulations [29,30,39] are suitable for exploiting sparsity, they come with a large set of inequality constraints, which makes them intractable on nearfuture quantum computers that are limited in the number of qubits. In this context, permutation matrices were shown to be a powerful tool for synchronization or shape matching [3,4,6]. In the following, we therefore propose a formulation based on assignment matrices that grows linearly in the number of required qubits for detections, tracks and frames.…”

Section: Quantum Motmentioning

confidence: 99%

“…Using this approach, a quantum speedup, which further scales with system size and temperature, has already been shown for applications in physics [35,36]. While there is a range of advantages that quantum computing can provide in the future, mapping a problem to an AQC is not trivial and often requires reformulating the problem from scratch, even for well investigated tasks [4,6]. On the one hand, the problem needs to be matched to the Ising model, on the other hand, real quantum computers have a very limited number of qubits and are still prone to noise, which requires tuning of the model to handle the limitations.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Adiabatic Quantum Computing for Multi Object Tracking

Zaech¹,

Liniger²,

Danelljan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Multi-Object Tracking (MOT) is most often approached in the tracking-by-detection paradigm, where object detections are associated through time. The association step naturally leads to discrete optimization problems. As these optimization problems are often NP-hard, they can only be solved exactly for small instances on current hardware. Adiabatic quantum computing (AQC) offers a solution for this, as it has the potential to provide a considerable speedup on a range of NP-hard optimization problems in the near future. However, current MOT formulations are unsuitable for quantum computing due to their scaling properties. In this work, we therefore propose the first MOT formulation designed to be solved with AQC. We employ an Ising model that represents the quantum mechanical system implemented on the AQC. We show that our approach is competitive compared with state-of-the-art optimizationbased approaches, even when using of-the-shelf integer programming solvers. Finally, we demonstrate that our MOT problem is already solvable on the current generation of real quantum computers for small examples, and analyze the properties of the measured solutions.

show abstract

Section: Related Workmentioning

confidence: 77%

Section: Quantum Motmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adiabatic Quantum Computing for Multi Object Tracking

Zaech¹,

Liniger²,

Danelljan³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…However, the limitation of quantum computers precluded them from solving large problems. To address this issue, instead of enforcing a penalty to QAP, Q-Match [67] was proposed to iteratively select and solve subproblems of QAP, which allows D-Wave annealers to efficiently deal with large problems. Another interesting work is Quan-tumSync [8], which addresses the synchronisation problem in the context of multi-image matching.…”

Section: Related Workmentioning

confidence: 99%

“…The above issues limit the scale of problems and quality of solutions attainable with current quantum annealers. However, quantum technology is advancing steadily, and the vision community should be prepared for potential breakthroughs, as like-minded colleagues are also advocating [7,8,19,37,65,67]. Moreover, our main algorithm combines quantum and classical computation to leverage the strengths of both paradigms.…”

Section: Practical Considerations and Limitationsmentioning

confidence: 99%

A Hybrid Quantum-Classical Algorithm for Robust Fitting

Doan¹,

Sasdelli²,

Suter³

et al. 2022

Preprint

View full text Add to dashboard Cite

Fitting geometric models onto outlier contaminated data is provably intractable. Many computer vision systems rely on random sampling heuristics to solve robust fitting, which do not provide optimality guarantees and error bounds. It is therefore critical to develop novel approaches that can bridge the gap between exact solutions that are costly, and fast heuristics that offer no quality assurances. In this paper, we propose a hybrid quantum-classical algorithm for robust fitting. Our core contribution is a novel robust fitting formulation that solves a sequence of integer programs and terminates with a global solution or an error bound. The combinatorial subproblems are amenable to a quantum annealer, which helps to tighten the bound efficiently. While our usage of quantum computing does not surmount the fundamental intractability of robust fitting, by providing error bounds our algorithm is a practical improvement over randomised heuristics. Moreover, our work represents a concrete application of quantum computing in computer vision. We present results obtained using an actual quantum computer (D-Wave Advantage) and via simulation 1 .

show abstract