Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

The Journal of Supercomputing

Hybrid quantum-classical 3D object detection using multi-channel quantum convolutional neural network

Emily Jimin Roh, Joo Yong Shim, Joongheon Kim, Soohyun Park

Published Feb 1, 2025• 9 citations
Publisher
DOI
10.1007/s11227-025-06968-7

Related papers

Hybrid quantum convolutional neural network for multi-channel image classification

Cai Zhang, Li Zheng, Haozhen Situ

Physica A Statistical Mechanics and its Applications•Nov 20, 2025•4 citations

Quantum Convolutional Neural Network Architectural Block for Object Detection

Zaid Aljundi, Adrian Turcanu

Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering•Jan 1, 2026•0 citations

Quantum-Eyes: Scalable Quantum Convolutional Neural Networks for Low-Overhead Object Detection [Research Frontier]

Joongheon Kim, Emily Jimin Roh, Chaemoon Im +1 more

Quantum neural networks (QNNs) are gaining attention as a promising foundation for next-generation machine learning. However, the practical deployment of QNNs, particularly in vision tasks, remains limited due to hardware constraints such as qubit scarcity. To address this challenge, this paper proposes a quantum convolutional neural network (QCNN)-based object detection framework designed for the noisy intermediate scale quantum (NISQ) era. The proposed method introduces four key components, patch processing, channel uploading, hybrid channel construction, and heterogeneous knowledge distillation, to enable efficient multi-channel object detection with a limited number of qubits. Furthermore, the knowledge distillation strategy facilitates the transfer of knowledge from a classical region proposal network (RPN) to its quantum counterpart, thereby enhancing detection accuracy despite the limitations of early-stage quantum machine learning. The evaluation results demonstrate the feasibility and efficiency of the proposed quantum object detection model, which can be effectively applied to complex vision tasks within existing quantum hardware limitations.

IEEE Computational Intelligence Magazine•Jul 14, 2025•2 citations

Fast Quantum Convolutional Neural Networks for Low-Complexity Object Detection in Autonomous Driving Applications

Emily Jimin Roh, Hankyul Baek, Donghyeon Kim +1 more

Object detection applications, especially in autonomous driving, have drawn attention due to the advancements in deep learning. Additionally, with continuous improvements in classical convolutional neural networks (CNNs), there has been a notable enhancement in both the efficiency and speed of these applications, making autonomous driving more reliable and effective. However, due to the exponentially rapid growth in the complexity and scale of visual signals used in object detection, there are limitations regarding computation speeds while conducting object detection solely with classical computing. Motivated by this, this paper proposes the quantum object detection engine (QODE), which implements a quantum version of CNN, named QCNN, in object detection. Furthermore, this paper proposes a novel fast quantum convolution algorithm that processes the multi-channel of visual signals based on a small number of qubits and constructs the output channel data, thereby achieving relieved computational complexity. Our QODE, equipped with fast quantum convolution, demonstrates feasibility in object detection with multi-channel data, addressing a limitation of current QCNNs due to the scarcity of qubits in the current era of quantum computing. Moreover, this paper introduces a heterogeneous knowledge distillation training algorithm that enhances the performance of our QODE.

IEEE Transactions on Mobile Computing•Sep 27, 2024•35 citations

Hybrid quantum-classical convolutional neural networks

Junhua Liu, Kwan Hui Lim, Kristin Lee Wood +3 more

arXiv:1911.02998•Science China Physics Mechanics and Astronomy•Aug 4, 2021•291 citations

Research on Object Detection Model Based on Quantum Convolutional Neural Network

Hui Gao

In recent years, object detection technology has undergone substantial evolution within the realms of deep learning and computer vision, particularly owing to the extensive utilization of traditional convolutional neural networks (CNNs). These technologies perform well in a variety of applications, such as autonomous driving, security monitoring, and medical image analysis. However, traditional CNNS face many challenges when dealing with complex scenes and small target detection, such as background interference, target scale change and occlusion. In this paper, a novel Hybrid model, Hybrid Quantum-Cascade FPN, is proposed, which cleverly combines the advantages of Quantum convolutional neural network (QCNN) and classical CNN. The design of Hybrid quantum-Cascade FPN model is divided into three key stages: first, the Quantum coarse screening stage, which quickly extracts low-level features through quantum feature coding; Next is the quantum-classical feature fusion stage, which uses the parallelism of quantum computing and the structural advantages of classical networks to carry out multi-scale feature fusion. The last is the classical fine-tuning stage, which further optimizes the model to improve the high-level semantic analysis ability. Experiments on multiple data sets show that the model performs well in both detection accuracy and inference speed, especially in complex background and small target detection tasks.

IEEE International Conference on Power, Intelligent Computing and Systems (ICPICS)•Aug 29, 2025•0 citations
Quantum neural networks (QNNs) are gaining attention as a promising foundation for next-generation machine learning. However, the practical deployment of QNNs, particularly in vision tasks, remains limited due to hardware constraints such as qubit scarcity. To address this challenge, this paper proposes a quantum convolutional neural network (QCNN)-based object detection framework designed for the noisy intermediate scale quantum (NISQ) era. The proposed method introduces four key components, patch processing, channel uploading, hybrid channel construction, and heterogeneous knowledge distillation, to enable efficient multi-channel object detection with a limited number of qubits. Furthermore, the knowledge distillation strategy facilitates the transfer of knowledge from a classical region proposal network (RPN) to its quantum counterpart, thereby enhancing detection accuracy despite the limitations of early-stage quantum machine learning. The evaluation results demonstrate the feasibility and efficiency of the proposed quantum object detection model, which can be effectively applied to complex vision tasks within existing quantum hardware limitations.
Object detection applications, especially in autonomous driving, have drawn attention due to the advancements in deep learning. Additionally, with continuous improvements in classical convolutional neural networks (CNNs), there has been a notable enhancement in both the efficiency and speed of these applications, making autonomous driving more reliable and effective. However, due to the exponentially rapid growth in the complexity and scale of visual signals used in object detection, there are limitations regarding computation speeds while conducting object detection solely with classical computing. Motivated by this, this paper proposes the quantum object detection engine (QODE), which implements a quantum version of CNN, named QCNN, in object detection. Furthermore, this paper proposes a novel fast quantum convolution algorithm that processes the multi-channel of visual signals based on a small number of qubits and constructs the output channel data, thereby achieving relieved computational complexity. Our QODE, equipped with fast quantum convolution, demonstrates feasibility in object detection with multi-channel data, addressing a limitation of current QCNNs due to the scarcity of qubits in the current era of quantum computing. Moreover, this paper introduces a heterogeneous knowledge distillation training algorithm that enhances the performance of our QODE.
In recent years, object detection technology has undergone substantial evolution within the realms of deep learning and computer vision, particularly owing to the extensive utilization of traditional convolutional neural networks (CNNs). These technologies perform well in a variety of applications, such as autonomous driving, security monitoring, and medical image analysis. However, traditional CNNS face many challenges when dealing with complex scenes and small target detection, such as background interference, target scale change and occlusion. In this paper, a novel Hybrid model, Hybrid Quantum-Cascade FPN, is proposed, which cleverly combines the advantages of Quantum convolutional neural network (QCNN) and classical CNN. The design of Hybrid quantum-Cascade FPN model is divided into three key stages: first, the Quantum coarse screening stage, which quickly extracts low-level features through quantum feature coding; Next is the quantum-classical feature fusion stage, which uses the parallelism of quantum computing and the structural advantages of classical networks to carry out multi-scale feature fusion. The last is the classical fine-tuning stage, which further optimizes the model to improve the high-level semantic analysis ability. Experiments on multiple data sets show that the model performs well in both detection accuracy and inference speed, especially in complex background and small target detection tasks.