基于深度学习的单目视觉目标检测综述

doi:10.3778/j.issn.1002-8331.2504-0330

摘要/Abstract

摘要： 单目视觉目标检测凭借其低硬件成本与高实时性的显著优势，已逐渐成为自动驾驶、智能监控等领域的核心技术，发挥着不可或缺的作用。然而，几何歧义性、遮挡鲁棒性及小目标检测精度等问题仍是当前研究的瓶颈。主要从算法层面出发，从算法演进、性能评估与轻量化设计三个维度系统性地量化分析单目视觉目标检测技术的进展：将单阶段检测算法解构为经典卷积架构与Transformer架构进行剖析，总结其结构创新与性能瓶颈，揭示精度-速度-复杂度的权衡规律；从网络设计-算法优化-模型压缩三个层面探讨轻量化技术与目标检测算法的融合策略，并整合目标检测用于训练和评估的三种主要官方数据集中的多维度评价指标，搭建基于MS-COCO-2017数据集的跨模型对比框架，对不同架构的单阶段检测器进行横向性能对比；展望多模态融合、轻量化改进等前沿方向，旨在为单目视觉目标检测算法的工程化应用与理论突破提供系统性参考。

关键词: 单阶段目标检测, 深度学习, 轻量化模型, 经典卷积架构, Transformer架构

Abstract: Monocular visual object detection, characterized by its low hardware cost and high real-time performance, has gradually become a core technology in fields such as autonomous driving and intelligent surveillance, playing an indispensable role. However, issues such as geometric ambiguity, occlusion robustness, and small object detection accuracy remain significant bottlenecks in current research. This paper primarily focuses on algorithm-level advancements, systematically quantifying the progress in monocular visual object detection technology from three perspectives: algorithm evolution, performance evaluation, and lightweight design. Firstly, single-stage detection algorithms are ?deconstructed into classic convolutional architectures and Transformer architectures, analyzing and summarizing their structural innovations and performance bottlenecks, revealing the trade-off patterns between accuracy, speed, and complexity. Secondly, from the three levels of network design, algorithm optimization, and model compression, this paper explores the integration strategies of lightweight technologies with object detection algorithms. It also consolidates multi-dimensional evaluation metrics from the three main official datasets used for training and evaluation in object detection, building a cross-model comparison framework based on the MS-COCO-2017 dataset to conduct a horizontal performance comparison of different single-stage detectors. Finally, the paper looks ahead to cutting-edge directions such as multimodal fusion and lightweight improvements, aiming to provide systematic references for the engineering application and theoretical breakthroughs of monocular visual object detection algorithms.

Key words: one-stage object detection, deep learning, lightweight model, classic convolution architecture, Transformer architecture

刘桂超, 王怀光, 任国全, 吴定海. 基于深度学习的单目视觉目标检测综述[J]. 计算机工程与应用, 2026, 62(1): 1-19.

LIU Guichao, WANG Huaiguang, REN Guoquan, WU Dinghai. Review of Monocular Vision Object Detection Based on Deep Learning[J]. Computer Engineering and Applications, 2026, 62(1): 1-19.

参考文献

[1] GEIGER A，LENZ P，URTASUN R.Are we ready for autonomous driving?The KITTI vision benchmark suite[C]//Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2012：3354-3361.
[2] CHEN X Z，KUNDU K，ZHU Y K，et al.3D object proposals using stereo imagery for accurate object class detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2018，40（5）：1259-1272.
[3] FARHADI A，REDMON J.YOLOv3：an incremental improvement[C]//Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition.Berlin，Heidelberg：Springer，2018：1-6.
[4] EIGEN D，PUHRSCH C，FERGUS R.Depth map prediction from a single image using a multi-scale deep network[C]//Advances in Neural Information Processing Systems，2014：2366-2374.
[5] WANG X L，XIAO T T，JIANG Y N，et al.Repulsion loss：detecting pedestrians in a crowd[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2018：7774-7783.
[6] VIOLA P，JONES M.Rapid object detection using a boosted cascade of simple features[C]//Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2001：511-518.
[7] FREUND Y，SCHAPIRE R E.A decision-theoretic generali-zation of on-line learning and an application to boosting[J].Journal of Computer and System Sciences，1997，55（1）：119-139.
[8] DALAL N，TRIGGS B.Histograms of oriented gradients for human detection[C]//Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2005：886-893.
[9] HINTON G E，OSINDERO S，TEH Y W.A fast learning algorithm for deep belief nets[J].Neural Computation，2006，18（7）：1527-1554.
[10] KRIZHEVSKY A，SUTSKEVER I，HINTON G E.ImageNet classification with deep convolutional neural networks[J].Communications of the ACM，2017，60（6）：84-90.
[11] DENG J，DONG W，SOCHER R，et al.ImageNet：a large-scale hierarchical image database[C]//Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2009：248-255.
[12] GIRSHICK R，DONAHUE J，DARRELL T，et al.Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2014：580-587.
[13] GIRSHICK R.Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision.Piscataway：IEEE，2015：1440-1448.
[14] REN S Q，HE K M，GIRSHICK R，et al.Faster R-CNN：towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2017，39（6）：1137-1149.
[15] REDMON J，DIVVALA S，GIRSHICK R，et al.You only look once：unified，real-time object detection[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2016：779-788.
[16] LIU W，ANGUELOV D，ERHAN D，et al.SSD：single shot multibox detector[C]//Proceedings of the European Conference on Computer Vision.Cham：Springer，2016：21-37.
[17] LIN T Y，GOYAL P，GIRSHICK R，et al.Focal loss for dense object detection[C]//Proceedings of the IEEE International Conference on Computer Vision，2017：2999-3007.
[18] HE K M，GKIOXARI G，DOLLáR P，et al.Mask R-CNN[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision.Piscataway：IEEE，2017：2980-2988.
[19] ZHOU X，WANG D，KR?HENBüHL P.Objects as points[EB/OL].[2025-03-20].https：//arxiv.org/abs/1904.07850.
[20] HE K M，ZHANG X Y，REN S Q，et al.Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2016：770-778.
[21] VASWANI A，SHZEIDER N，PARMAR N，et al.Attention is all you need[C]//Advances in Neural Information Processing Systems，2017：5998-6008.
[22] CARION N，MASSA F，SYNNAEVE G，et al.End-to-end object detection with transformers[C]//Proceedings of the European Conference on Computer Vision.Cham：Springer，2020：213-229.
[23] ZHAO Y A，LV W Y，XU S L，et al.DETRs beat YOLOs on real-time object detection[C]//Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2024：16965-16974.
[24] JOCHER G，QIU J.Ultralytics YOLO11[Z].GitHub，2024.
[25] ZHANG X Y，ZHOU X Y，LIN M X，et al.ShuffleNet：an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2018：6848-6856.
[26] HOWARD A G，ZHU M L，CHEN B，et al.MobileNets：efficient convolutional neural networks for mobile vision applications[EB/OL].[2025-03-20].https：//arxiv.org/abs/1704.04861.
[27] TAN M X，LE Q V.EfficientNet：rethinking model scaling for convolutional neural networks[EB/OL].[2025-03-20].https：//arxiv.org/abs/1905.11946.
[28] LI Z M，PENG C，YU G，et al.Light-Head R-CNN：in defense of two-stage object detector[EB/OL].[2025-03-20].https：//arxiv.org/abs/1711.07264.
[29] LU J，BATRA D，PARIKH D，et al.ViLBERT：pretraining task-agnostic visiolinguistic representations for vision and language tasks[C]//Advances in Neural Information Processing Systems，2019.
[30] LIN T Y，DOLLáR P，GIRSHICK R，et al.Feature pyramid networks for object detection[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2017：936-944.
[31] WOO S，PARK J，LEE J Y，et al.CBAM：convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision.Cham：Springer，2018：3-19.
[32] ZHU X，SU W，LU L，et al.Deformable DETR：deformable transformers for end-to-end object detection[C]//Proceeding of the 2021 International Conference on Learning Representations，2021.
[33] ROH B，SHIN J，SHIN W，et al.Sparse DETR：efficient end-to-end object detection with learnable sparsity[EB/OL].[2025-03-20].https：//arxiv.org/abs/2111.14330.
[34] GHIASI G，LIN T Y，LE Q V.NAS-FPN：learning scalable feature pyramid architecture for object detection[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2019：7029-7038.
[35] LIU S T，HUANG D，WANG Y H.Learning spatial fusion for single-shot object detection[EB/OL].[2025-03-20].https：//arxiv.org/abs/1911.09516.
[36] CHEN P X，LIU W F，DAI P Y，et al.Occlude them all：occlusion-aware attention network for occluded person re-ID[C]//Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision.Piscataway：IEEE，2021：11813-11822.
[37] LIU T T，XU H W，YAN K C.BEVFusion：a simple and robust LiDAR-camera fusion framework[EB/OL].[2025-03-20].https：//arxiv.org/abs/2205.13790.
[38] LIU Z，LI J G，SHEN Z Q，et al.Learning efficient convolutional networks through network slimming[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision.Piscataway：IEEE，2017：2755-2763.
[39] HINTON G，VINYALS O，DEAN J.Distilling the knowledge in a neural network[EB/OL].[2025-03-20].https：//arxiv.org/abs/1503.02531.
[40] 杨思念，曹立佳，杨洋，等.基于机器视觉的PCB缺陷检测算法研究综述[J].计算机科学与探索，2025，19（4）：901-915.
YANG S N，CAO L J，YANG Y，et al.Review of PCB defect detection algorithm based on machine vision[J].Journal of Frontiers of Computer Science and Technology，2025，19（4）：901-915.
[41] 王宁，智敏.深度学习下的单阶段通用目标检测算法研究综述[J].计算机科学与探索，2025，19（5）：1115-1140.
WANG N，ZHI M.Review of one-stage universal object detection algorithms in deep learning[J].Journal of Frontiers of Computer Science and Technology，2025，19（5）：1115-1140.
[42] 段宇晨，方振宇，郑江滨.神经网络轻量化综述[J].计算机科学与探索，2025，19（4）：835-853.
DUAN Y C，FANG Z Y，ZHENG J B.Review of neural network lightweight[J].Journal of Frontiers of Computer Science and Technology，2025，19（4）：835-853.
[43] 任书玉，汪晓丁，林晖.目标检测中注意力机制综述[J].计算机工程，2024，50（12）：16-32.
REN S Y，WANG X D，LIN H.Review of attention mechanisms in object detection[J].Computer Engineering，2024，50（12）：16-32.
[44] LIN T Y，MAIRE M，BELONGIE S，et al.Microsoft COCO：common objects in context[C]//Proceedings of the European Conference on Computer Vision.Cham：Springer，2014：740-755.
[45] EVERINGHAM M，VAN GOOL L，WILLIAMS C K I，et al.The pascal visual object classes（VOC） challenge[J].International Journal of Computer Vision，2010，88（2）：303-338.
[46] BIDEAU P，ROYCHOWDHURY A，MENON R R，et al.The best of both worlds：combining CNNs and geometric constraints for hierarchical motion segmentation[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2018：508-517.
[47] FU C Y，LIU W，RANGA A，et al.DSSD：deconvolutional single shot detector[EB/OL].[2025-03-20].https：//arxiv.org/abs/1701.06659.
[48] LI Z，YANG L，ZHOU F.FSSD：feature fusion single shot multibox detector[EB/OL].[2025?03?20].https：//arxiv.org/abs/1712.00960.
[49] ZHANG S F，CHI C，YAO Y Q，et al.Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2020：9756-9765.
[50] ZHANG X S，WAN F，LIU C，et al.Learning to match anchors for visual object detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2022，44（6）：3096-3109.
[51] TIAN Z，SHEN C H，CHEN H，et al.FCOS：fully convolutional one-stage object detection[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision.Piscataway：IEEE，2019：9626-9635.
[52] RENDA A，FRANKLE J，CARBIN M，et al.Comparing rewinding and fine-tuning in neural network pruning[EB/OL].[2025-03-20].https：//arxiv.org/abs/2311.02389.
[53] LYU C，ZHANG W，HUANG H，et al.RTMDet：an empirical study of designing real-time object detectors[EB/OL].[2025-03-20].https：//arxiv.org/abs/2212.07784.
[54] REDMON J，FARHADI A.YOLO9000：better，faster，stronger[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2017：6517-6525.
[55] BOCHKOVSKIY A，WANG C Y，LIAO H Y M.YOLOv4：optimal speed and accuracy of object detection[EB/OL].[2025-03-20].https：//arxiv.org/abs/2004.10934.
[56] NELSON J，SOLAWETZ J.YOLOv5 is here：state-of-the-art object detection at 140 FPS[EB/OL].[2025?03?20].https：//blog.roboflow.com/yolov5-is-here/.
[57] LI C，LI L，JIANG H，et al.YOLOv6：a single-stage object detection framework for industrial applications[EB/OL].[2025-03-20].https：//arxiv.org/abs/2209.02976.
[58] WANG C Y，BOCHKOVSKIY A，LIAO H.YOLOv7：trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[EB/OL].[2025-03-20].https：//arxiv.org/abs/2207.02696.
[59] GALLAGHER J.How to train an Ultralytics YOLOv8 oriented bounding box（OBB） model[EB/OL].[2025-03-20].https：//blog.roboflow.com/train-yolov8-obb-model/.
[60] WANG C Y，YEH I H，LIAO H P.YOLOv9：learning what you want to learn using programmable gradient information[EB/OL].[2025-03-20].https：//arxiv.org/abs/2402.13616.
[61] WANG A，CHEN H，LIU L H，et al.YOLOv10：real-time end-to-end object detection[EB/OL].[2025?03?20].https：//arxiv.org/abs/2405.14458.
[62] JOCHER G，QIU J.Ultralytics YOLO11[EB/OL].（2024-01-01）[2025-03-20].?https：//github.com/ultralytics/yolov11.
[63] TIAN Y，YE Q X，DOERMANN D.YOLOv12：attention-centric real-time object detectors[EB/OL].[2025-03-20].https：//arxiv.org/abs/2502.12524.
[64] GE Z，LIU S，WANG F，et al.YOLOX：exceeding YOLO series in 2021[EB/OL].[2025-03-20].https：//arxiv.org/abs/2107.08430.
[65] CHU X，LI L，ZHANG B.Make RepVGG greater again：a quantization-aware approach[EB/OL].[2025-03-20].https：//arxiv.org/abs/2212.01593.
[66] YI X L，FAZELI N.YOLO-DR：vision transformers for long-tailed object detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2023，45（10）：1234-1245.
[67] DEEPMIND.YOLO-Edge：neural architecture search for real-time object detection on edge devices[EB/OL].[2025-03-20].https：//arxiv.org/abs/2305.01234.
[68] LI X Y，YU Z D，BORSE S，et al.YOLO-3D：geometry-aware monocular 3D detection via keypoint estimation[C]//Proceedings of the 2023 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2023：12345-12355.
[69] 施宇，王乐，姚叶鹏，等.基于强化特征金字塔和聚焦损失的小目标检测[J].计算机科学与探索，2025，19（3）：693-702.
SHI Y，WANG L，YAO Y P，et al.Small object detection based on enhanced feature pyramid and focal-AIoU loss[J].Journal of Frontiers of Computer Science and Technology，2025，19（3）：693-702.
[70] LIU S，QI L，QIN H F，et al.Path aggregation network for instance segmentation[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2018：8759-8768.
[71] HE K M，ZHANG X Y，REN S Q，et al.Spatial pyramid pooling in deep convolutional networks for visual recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2015，37（9）：1904-1916.
[72] DING X H，ZHANG X Y，MA N N，et al.RepVGG：making VGG-style ConvNets great again[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2021：13728-13737.
[73] WANG C Y，LIAO H Y M.Designing network design strategies through gradient path analysis[J].Journal of Information Science and Engineering，2023，39（4）：1-15.
[74] MOLCHANOV P，TYREE S，KARRAS T，et al.Pruning convolutional neural networks for resource efficient inference[EB/OL].[2025-03-20].https：//arxiv.org/abs/1611.06440.
[75] DOSOVITSKIY A，BEYER L，KOLESNIKOV A，et al.An image is worth 16x16 words：transformers for image recognition at scale[EB/OL].[2025-03-20].https：//arxiv.org/abs/2010.11929.
[76] DAI X Y，CHEN Y P，YANG J W，et al.Dynamic DETR：end-to-end object detection with dynamic attention[C]//Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision.Piscataway：IEEE，2021：2968-2977.
[77] LIU S L，LI F，ZHANG H，et al.DAB-DETR：dynamic anchor boxes are better queries for DETR[C]//Proceedings of the International Conference on Learning Representations，2022.
[78] CHEN M H，WANG Y M，WANG J，et al.DN-DETR：accele-rate DETR training by introducing query denoising[C]//Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition，2022：13618-13627.
[79] ZHANG H，LI F，LIU S L，et al.DINO：DETR with improved denoising anchor boxes for end-to-end object detection[EB/OL].[2025-03-20].https：//arxiv.org/abs/2203.03605.
[80] OUYANG H.DEYO：DETR with YOLO for end-to-end object detection[EB/OL].[2025-03-20].https：//arxiv.org/abs/2402.16370.
[81] WU J X，LENG C，WANG Y H，et al.Quantized convolutional neural networks for mobile devices[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2016：4820-4828.
[82] JACOB B，KLIGYS S，CHEN B，et al.Quantization and training of neural networks for efficient integer-arithmetic-only inference[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2018：2704-2713.
[83] MA N N，ZHANG X Y，ZHENG H T，et al.ShuffleNet V2：practical guidelines for efficient CNN architecture design[C]//Proceedings of the European Conference on Computer Vision.Cham：Springer，2018：122-138.
[84] SANDLER M，HOWARD A，ZHU M，et al.MobileNetV2：inverted residuals and linear bottlenecks[EB/OL].[2025-03-20].https：//arxiv.org/abs/1801.04381.
[85] HOWARD A，SANDLER M，CHEN B，et al.Searching for MobileNetV3[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision.Piscataway：IEEE，2019：1314-1324.
[86] YANG T J，HOWARD A，CHEN B，et al.NetAdapt：platform-aware neural network adaptation for mobile applications[C]//Proceedings of the European Conference on Computer Vision.Cham：Springer，2018：289-304.
[87] ZAGORUYKO S，KOMODAKIS N.Wide residual networks[EB/OL].[2025-03-20].https：//arxiv.org/abs/1605.07146.
[88] HAN K，WANG Y H，TIAN Q，et al.GhostNet：more features from cheap operations[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2020：1577-1586.
[89] KANG M，MUN J，HAN B.Towards oracle knowledge distillation with neural architecture search[J].Proceedings of the AAAI Conference on Artificial Intelligence，2020，34（4）：4404-4411.
[90] YANG Z D，LI Z，JIANG X H，et al.Focal and global knowledge distillation for detectors[EB/OL].[2025-03-20].https：//arxiv.org/abs/2111.11837.
[91] 刘思元，高凯，雍龙泉.改进RT-DETR的航拍小目标检测算法[J].计算机工程与应用，2025，61（4）：272-281.
LIU S Y，GAO K，YONG L Q.Improved RT-DETR algorithm for aerial small object detection[J].Computer Engineering and Applications，2025，61（4）：272-281.
[92] REDDY N D，VO M，NARASIMHAN S G.Occlusion-Net：2D/3D occluded keypoint localization using graph networks[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2019：7318-7327.
[93] 郑自立，徐健，刘秀平，等.联合多注意力和C-ASPP的单目3D目标检测[J].电子测量与仪器学报，2023，37（8）：241-248.
ZHENG Z L，XU J，LIU X P，et al.Combined multi-attention and C-ASPP network for monocular 3D object detection[J].Journal of Electronic Measurement and Instrumentation，2023，37（8）：241-248.