The DOI matches its Crossref record; 29 of the 29 references it cites were found in a registry or an index.
DeepVO: Towards end-to-end visual odometry with deep Recurrent Convolutional Neural Networks
Open accessAbstract
This paper studies monocular visual odometry (VO) problem. Most of existing VO algorithms are developed under a standard pipeline including feature extraction, feature matching, motion estimation, local optimisation, etc. Although some of them have demonstrated superior performance, they usually need to be carefully designed and specifically fine-tuned to work well in different environments. Some prior knowledge is also required to recover an absolute scale for monocular VO. This paper presents a novel end-to-end framework for monocular VO by using deep Recurrent Convolutional Neural Networks (RCNNs). Since it is trained and deployed in an end-to-end manner, it infers poses directly from a sequence of raw RGB images (videos) without adopting any module in the conventional VO pipeline. Based on the RCNNs, it not only automatically learns effective feature representation for the VO problem through Convolutional Neural Networks, but also implicitly models sequential dynamics and relations using deep Recurrent Neural Networks. Extensive experiments on the KITTI VO dataset show competitive performance to state-of-the-art methods, verifying that the end-to-end Deep Learning technique can be a viable complement to the traditional VO systems.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“More surprisingly, they show that the welltrained networks enable scale-consistent predictions over long videos, while the accuracy is still inferior to traditional methods because of ignoring geometric information.”
From DF-VO: What Should Be Learnt for Visual Odometry? (2021) · cites this paper as Wang et al. 2017
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“On the opposing side, end-to-end approaches can estimate the 6 degrees of freedom (DoF) poses directly from a sequence of RGB images, that is, it does not need engineering effort to design the modules in the traditional VO pipeline [4].”
From Transformer-Based Model for Monocular Visual Odometry: A Video Understanding Approach (2023) · cites this paper as [4]
The cited source says what the claim attributes to it.
In the citing paper
“This problem is called monocular visual odometry and often relies on geometric approaches that require considerable engineering effort for a specific scenario.”
From Transformer-Based Model for Monocular Visual Odometry: A Video Understanding Approach (2023) · cites this paper as [4]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“the amount of crater detections that ODN-detector produces is robust to disturbances or modifications of the image quality due to Gaussian noise, shot noise, brightness levels, as well as different illumination conditions.”
From Optical navigation for Lunar landing based on Convolutional Neural Network crater detector (2022) · cites this paper as [16]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 1×
“There have been some popular and powerful DNN architectures, such as VGGNet [22] and GoogLeNet [23], developed for computer vision tasks, producing remarkable performance.”
MethodGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
Are we ready for autonomous driving? The KITTI vision benchmark suite
Geiger, A., Lenz, P., & Urtasun, R. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6248074 · Cited by 14,897 · cited 2×
“In this section, we evaluate the proposed end-to-end monocular VO approach on the well-known KITTI VO/SLAM benchmark [3].”
BackgroundMethod(2016) · Cited by 11,679
ORB-SLAM: A Versatile and Accurate Monocular SLAM System
Mur-Artal, R., Montiel, J. M. M., & Tardos, J. D. (2015) · IEEE Transactions on Robotics · doi:10.1109/tro.2015.2463671 · Cited by 6,778 · cited 1×
“Examples include keyframe base PTAM [9] and ORB-SLAM [10].”
BackgroundParallel Tracking and Mapping for Small AR Workspaces
Klein, G., & Murray, D. (2007) · 2007 6th IEEE and ACM International Symposium on Mixed and Augmented Reality · doi:10.1109/ismar.2007.4538852 · Cited by 4,346 · cited 1×
“Examples include keyframe base PTAM [9] and ORB-SLAM [10].”
BackgroundFlownet: Learning optical flow with convolutional networks
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., Smagt, P. V. D., Cremers, D., & Brox, T. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.316 · Cited by 4,313 · cited 2×
“In order to reduce both the training time and data required to converge, the CNN is based on a pre-trained FlowNet model [24].”
MethodMonoSLAM: Real-time single camera SLAM
Davison, A. J., Reid, I. D., Molton, N. D., & Stasse, O. (2007) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2007.1049 · Cited by 3,959 · cited 1×
“To mitigate this problem, visual Simultaneous Localisation and Mapping (SLAM) or Structure from Motion (SfM) can be adopted to maintain a feature map for drift correction along with pose estimation [8].”
MethodORB-SLAM: A Versatile and Accurate Monocular SLAM System
Mur-Artal, R., Montiel, J. M. M., & Tardos, J. D. (2015) · Zaguan (Universidad de Zaragoza) · Cited by 3,153
Multiple view geometry in computer vision
Opower, H. (2002) · Optics and Lasers in Engineering · doi:10.1016/s0143-8166(01)00145-2 · Cited by 3,044
Flagged first, then by citations
Research works printed but not linked: 1
- zaremba, 2014, Learning to execute
Cited by
Sorted by citations
Items 801–900 of 969
FRSK visual odometey based on FREAK
Chen, Z., & Chen, J. (2021) · International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE) · doi:10.1109/aemcse51986.2021.00221
Fully Differentiable and Interpretable Model for VIO with 4 Trainable ParametersPreprint
Chen, Z., Du, H., Liao, Y., Wang, Y., & Xiong, R. (2021) · arXiv (Cornell University)
Improving GNSS Positioning using Neural Network-based CorrectionsPreprint
Kanhere, A. V., Gupta, S., Shetty, A., & Gao, G. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.09581
“INTRODUCTION In the last decade, deep learning has been applied in several localization applications involving complex and high-dimensional sensor inputs, such as camera images and LiDAR pointclouds [Kendall et al., 2015; Wang et al., 2017; Choy et al., 2020].”
Neural Network for 2D Range Scanner Navigation System
Spampinato, G., Bruna, A. R., Guarneri, I., & Giacalone, D. (2021) · Advances in Science Technology and Engineering Systems Journal · doi:10.25046/aj060539
Discrete Spherical Image Representation for CNN‐Based Spherical Camera Pose Estimation
Wang, L., & Li, S. (2021) · IEEJ Transactions on Electrical and Electronic Engineering · doi:10.1002/tee.23506
Firintepe, A., Habib, S., Pagani, A., & Stricker, D. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-90739-6_6
“Camera pose tracking is an intrinsic part of visual odometry, where several works have focused on CNN and LSTM combinations for pose tracking [42, 36, 8].”
BackgroundLearning to Detect Geometric Structures from Images for 3D Parsing
Zhou, Y. (2020) · eScholarship (California Digital Library)
Research on SLAM Algorithm of Mobile Robot Vision Based on Deep Learning
Song, H., Ren, B., Zhao, Z., & Cui, J. (2021) · Global Reliability and Prognostics and Health Management Conference (PHM) · doi:10.1109/phm-nanjing52125.2021.9612660
Robust Vision-based Simultaneous Localization and Mapping for Highly Dynamic Scenes
Zhang, Z., Lei, Q., Li, C., Zhuang, Z., & Yan, B. (2021) · International Conference on UK-China Emerging Technologies (UCET) · doi:10.1109/ucet54125.2021.9674991
RAM-VO: A Recurrent Attentional Model for Visual Odometry
Cleveston, I., & Colombini, E. L. (2021) · doi:10.5753/wtdr_ctdr.2021.18684
“In 2017, Wang et al. [Wang et al. 2017] proposed the DeepVO, which is an end-to-end monocular architecture capable of extracting visual features directly from the input raw images with a CNN and determining temporal relation with an LSTM.”
BackgroundDiagnosing Deep SLAM for Domain-Shift Localization
Yoshida, M., & Kanji, T. (2022) · IEEE/SICE International Symposium on System Integration (SII) · doi:10.1109/sii52469.2022.9708872
Parameter-adaptive multi-frame joint pose optimization method
Li, S., Xian, Y., Wu, W., Zhang, T., & Li, B. (2022) · The Visual Computer · doi:10.1007/s00371-022-02476-4
Bruno, H. M. S. (2020) · doi:10.47749/t/unicamp.2020.1129339
Kweon, I. S., & Rameau, F. (2021) · Springer eBooks · doi:10.1007/978-3-030-63416-2_848
Attention-Based Deep Odometry Estimation on Point Clouds
Kapoor, P., Nowruzi, F. E., Kolhatkar, D., & Laganière, R. (2022) · SN Computer Science · doi:10.1007/s42979-022-01212-y
“DeepVO [16] proposes a model architectures aimed at using deep learning to solve the VO problem in an endto-end fashion.”
da Silva, A. M. R., Casqueiro, G., Angonese, A. T., & Rosa, P. F. F. (2022) · Communications in computer and information science · doi:10.1007/978-3-031-08443-0_3
Monocular Depth Estimation of Indoor Environments Using Enhanced Rgb-D DatasetPreprint
Kim, J., Shin, Y.-G., & Chung, M. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4172041
Ddm-Vslam: Depth Prediction for Dynamic Culling Monocular Vision SlamPreprint
Bai, Z., Liu, X., Zhou, M., & Woźniak, M. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4195239
Deep Learned Ground Penetrating Radar Subsurface Features for Robot LocalizationPreprint
Wickramanayake, S., Thiyagarajan, K., & Kodagoda, S. (2022) · doi:10.36227/techrxiv.20618097
Deep Learned Ground Penetrating Radar Subsurface Features for Robot LocalizationPreprint
Wickramanayake, S., Thiyagarajan, K., & Kodagoda, S. (2022) · doi:10.36227/techrxiv.20618097.v1
A new visual odometry algorithm based on multi-path deep fully convolutional neural networks
Chen, B., Yan, K., Cao, R., Zhang, T., & Zhang, X. (2022) · International Symposium on Computer Technology and Information Science (ISCTIS) · doi:10.1117/12.2653846
“Similar to the train/test split in DeepVO [23] and ESPVO [24], We utilize the Sequence 00, 01, 02, 03, 08, 09 for training, and the Sequence 04, 05, 06, 07, 10 for evaluation.”
MethodTu, C.-H., & Lu, E. H.-C. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-21743-2_4
Backpropagation Based Deep Multispectral VO Drift Elimination
Khan, A. A. N., & Aouf, N. (2022) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta56598.2022.10034568
“Since the introduction of DeepVO [ 7 ], the field has witnessed an ever-expanding stream of End-to-End models come to fruition.”
BackgroundMethodCooperative Eye Tracking and Localization Based on Corneal Imaging
Kanai, Y., & Takemura, K. (2023) · IEEE/SICE International Symposium on System Integration (SII) · doi:10.1109/sii55687.2023.10039095
“The relative displacement O was computed by DeepVO [14] rather than using the IMU.”
MethodAn End-to-End Visual Odometry System Based On Multi-Branch Convolutional Neural Network
Wei, Z., Wang, L., Pan, W., Chen, Z., & Li, Z. (2022) · International Conference on Electronic Information Technology and Computer Engineering (EITCE) · doi:10.1145/3573428.3573756
“Sequence VISO2-M [18] DeepVO [4] CGC-ULSAM [12] Pure-CNN-based Method Proposed”
BackgroundMethodInfluentialResearches Advanced in Visual Simultaneously Localization and Mapping based on Deep Learning
Hao, H. (2022) · International Conference on Robotics, Intelligent Control and Artificial Intelligence (RICAI) · doi:10.1145/3584376.3584595 · Cited by 1
“focused on the broad topic of DeepVO, first providing a framework based on RCNN for monocular VO [13], and then, the following year, proposing ESP-VO framework based on a deep recurrent convolutional neural network framework.”
MethodA Position Correction Model for AUV Navigation with Sequential Learning-Assisted State Estimation
Zhang, X., Qiu, Y., Ohya, A., Yorozu, A., & He, B. (2023) · IEEE Underwater Technology (UT) · doi:10.1109/ut49729.2023.10103372
“proposed an endto-end framework based on Recurrent Convolutional Neural Networks (RCNNs) for Visual Odometry (VO) problem [6].”
MethodExploring Self-Attention for Visual Intersection Classification
Nakata, H., Tanaka, K., & Takeda, K. (2023) · Journal of Advanced Computational Intelligence and Intelligent Informatics · doi:10.20965/jaciii.2023.p0386
Performance Comparison of Different Visual Odometry Architectures to Improve ATDN vSLAM
Bogár, G. R., Szántó, M., & Szemenyei, M. (2022) · European Conference on Electrical Engineering and Computer Science (EECS) · doi:10.1109/elecs55825.2022.00035
“Previous results [30], and the works discussed below [31]–[34]) managed to show that scale-correct results can be assured through supervised learning – in contrast with the traditional VO methods, which did not manage to solve scale-ambiguity without the introduction of various scale-defining inputs, such as dimension…”
BackgroundResultBilal, M., Hanif, M. S., Munawar, K., & Al-Saggaf, U. M. (2023) · IEEE Access · doi:10.1109/access.2023.3284463
“Recurrent convolutional neural networks are employed to estimate pose from color images in monocular settings in [12].”
BackgroundMethodInfluentialGolroudbari, A. A., & Sabour, M. H. (2023) · doi:10.22541/au.168727976.63672670/v1 · Cited by 1
AdVLO: Region Selection via Attention-Driven for Visual LiDAR Odometry
Lam, H., Pho, K., & Yoshitaka, A. (2023) · Lecture notes in computer science · doi:10.1007/978-981-99-5834-4_7
Shin, J., Jeon, M., & Kim, A. (2023) · The Journal of Korea Robotics Society · doi:10.7746/jkros.2023.18.3.260
Machine Learning Components for Autonomous Navigation Systems
Samal, K., & Wolf, M. (2023) · Springer eBooks · doi:10.1007/978-3-031-40677-5_9
HO3-SLAM: Human-Object Occlusion Ordering SLAM Framework for Traversability Prediction
Liang, J. T. Y., & Kanji, T. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-47634-1_25
Monocular visual odometry based on epipolar geometry
da Silva, R. S. (2021) · doi:10.47749/t/unicamp.2021.1164461
Research on visual SLAM system based on near-infrared perception
Liu, M., Li, B., Qian, X., Ma, R., Zhu, P., Liao, C., Tian, H., Li, X., Wang, X., & Li, X. (2023) · Optoelectronic Imaging and Multimedia Technology · doi:10.1117/12.2689364
Motion Consistency Loss for Monocular Visual Odometry with Attention-Based Deep Learning
Françani, A. O., & Máximo, M. R. O. A. (2023) · Latin American Robotics Symposium, Brazilian Symposium on Robotics and Workshop on Robotics in Education (LARS/SBR/WRE) · doi:10.1109/lars/sbr/wre59448.2023.10332921 · Cited by 1
“…the 6-DoF, one can rely on traditional geometry-based approaches [9], [10], deep learning-based approaches through end-to-end architectures [11], [12], and hybrid methods that mix deep learning-based and geometry-based algorithms depending on the component of the visual…”
BackgroundMethodAn Attention-Based Odometry Framework for Multisensory Unmanned Ground Vehicles (UGVs)
Xiao, Z., & Zhang, G. (2023) · Drones · doi:10.3390/drones7120699
“DeepVO [28] DeepLO [29] Au et al. [30] Chen et al. [31] Ema-Vio [32] DeepLIO [33] LOAM [”
BackgroundMethodInfluentialSelf-Supervised Sparse Direct Visual Odometry with Half-Geometric Correspondence Network
Zhang, T., Li, Q., Zhou, M., & Yu, F. R. (2024) · Frontiers in artificial intelligence and applications · doi:10.3233/faia231259
Tejaswi, K., Lee, T., & Kang, C.-K. (2024) · AIAA SciTech Forum · doi:10.2514/6.2024-0948
SV-HaRF: a scene representation method for sparse image view synthesis
Liu, W., Zhang, Y., Zou, X., Ma, R., & Tao, P. (2024) · International Conference on Computer Graphics, Artificial Intelligence, and Data Processing (ICCAID) · doi:10.1117/12.3026707
“DeepVO [5] uses a deep recursive convolutional neural network (RCNN) to directly infer poses from a series of raw RGB images or videos.”
MethodVirtual Vectors for Vision-Based Simultaneous Localization and Mapping System
Cui, J., Huang, Y., Luo, X., & Bai, Y. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3412635
Wang, Y., Zhang, Y., Liu, X., & Zhou, Z. (2024) · International Conference on Advanced Algorithms and Control Engineering (ICAACE) · doi:10.1109/icaace61206.2024.10549600
“In 2017, Wang et al. proposed Deep-VO [5].”
BackgroundEnhancing Direct Visual Odometry with Deblurring and Saliency Maps
Gjerde, M. K., Nasrollahi, K., Moeslund, T. B., & Haurum, J. B. (2024) · doi:10.1145/3653946.3653970 · Cited by 1
“[8] In 2021, Guo et al. . presented DeblurSLAM [24].”
BackgroundA Mixed-Precision DNN Compiler for Robotic Computing on FPGA
Li, Q., Huang, H., Shen, A., Li, K., Huang, S., & Yu, H. (2023) · International Conference on Robotics, Automation and Artificial Intelligence (RAAI) · doi:10.1109/raai59955.2023.10601260
“For visual odometry, there are models like DeepVo [6] and UndeepVo [7] whereas for human pose estimation networks there are OpenPose [8], HRNet [9], Fig.”
MethodVisual SLAM Based on Improved Line Filtering Decision and Weight Optimization
Cao, Y., Luo, Z., & Deng, Z. (2024) · IEEE Access · doi:10.1109/access.2024.3481420
“For deep learning-based visual SLAM, such as DeepVO [11] and CNN-SLAM [12], they improve adaptability to complex scenes and changes but require training on large-scale datasets and may produce inexplicable results in some scenarios.”
BackgroundResearch on Key Technologies for Analyzing Quantum Computing with Deep Learning
Chen, K., Shi, L., He, C., Yu, Y., & Deng, K. (2024) · SPIN · doi:10.1142/s2010324724400083
金, 学., & 刘, 嵩. (2024) · Réngōng zhìnéng qiányán yǔ yìngyòng · doi:10.52810/faai.2024.002
Safety Assurance for Autonomous Systems with Multiple Sensor Modalities
Balakrishnan, A., Bernard, R., Narayanan, S., Kudalkar, V., Zhao, Y.-Q., Nagaraja, P., Markov, G., Budnik, C., Degen, H., Lindemann, L., et al. (2024) · International Conference on Formal Methods and Models for Co-Design · doi:10.1109/memocode63347.2024.00018 · Cited by 1
Dynamic Object Suppression in Visual Odometry via Adaptive Masked Flow Refinement
Ling, D., Xu, L. J., Ren, X., & Wang, X. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-8792-0_16
Chen, S., Kong, D., Lu, L., Chen, D., & Chen, D. (2024) · Measurement Science and Technology · doi:10.1088/1361-6501/ad976b
“DeepVO [25] and ESP-VO [26], stand as seminal achievements in learning-based VO.”
Huang, M. J., Lu, S., & Zhuo, G. (2024) · SAE technical papers on CD-ROM/SAE technical paper series · doi:10.4271/2024-01-7031
Fast Online Adaptation of Visual SLAM via Variational Information Transfer and Preservation
Xu, S., Xiong, H., Wu, Q., Wang, Z., Berkovsky, S., & Wang, Z. (2024) · ACM International Conference on Multimedia in Asia (MMAsia) · doi:10.1145/3696409.3700212
Overview of the current development in Visual-Inertial Systems
Wei, M., & Meng, Q. (2024) · Progress in Medical Devices · doi:10.61189/521889ygxkmc
Atia, M. M. (2025) · doi:10.1002/9781394155293.ch12
A Monocular Visual Odometry Model Based on Transformer Using Shifted Windows
Li, J., Li, J., Chen, W., Xue, T., & Zhang, J. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-0777-8_23
Semantic-Aware Monocular Point-Line Visual Odometry in Dynamic Environments
Wen, H., Chen, X., Xu, L. J., Ren, X., & Wang, X. (2024) · China Automation Congress (CAC) · doi:10.1109/cac63892.2024.10865087
Self-citationVisual SLAM in Underground Environments: Preliminary Results
Federico, B., Pablo, M., & Llofriu, M. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-80366-6_18
Thermal SLAM: Harnessing Temperature Variations to Enhance Object Detection and Tracking Performance
Salem, F. G. I., & Gedık, O. S. (2025) · Black Sea Journal of Engineering and Science · doi:10.34248/bsengineering.1578563
Transformer-Based Visual Odometry from RGB-D information
Zhou, Y., Zhou, Z., Jiao, R., Tao, Z., & Lv, J. H. (2024) · International Conference on Artificial Intelligence, Robotics, and Communication (ICAIRC) · doi:10.1109/icairc64177.2024.10899910
“DeepVO [2] developed an end-to-end framework using deep recurrent convolutional neural networks (RCNNs) to estimate camera poses directly from RGB sequences, bypassing traditional modules.”
Optimization‐Based Geometric Enhancement and Motion Estimation for Non‐Cooperative Spacecrafts
Zhang, C., Yu, H., Liang, Q., & Peng, J. (2025) · Journal of Field Robotics · doi:10.1002/rob.22540
“The former uses an end ‐ to ‐ end network but lacks interpretability (Wang et al. 2017).”
LN-SLAM: deep learning-based underground low-light loop closure SLAM
Wu, Y., Li, J., Chang, Q., Wang, X., Han, L., Zhang, H., & Duan, Y. (2025) · International Conference on Physics, Photonics, and Optical Engineering (ICPPOE) · doi:10.1117/12.3060554
Autonomous Driving System Based on Deep Q-Learning: A Survey of Attacks and Defenses
Gupta, R., Arora, N., Kangan, Ottalwar, R., Upadhyay, K., & Jayaprakash, L. (2025) · Communications in computer and information science · doi:10.1007/978-3-031-84394-5_4
Qi, F., & Ni, X. (2024) · IEEE International Conference on Frontiers Technology of Information and Computer (ICFTIC) · doi:10.1109/icftic64248.2024.10913162
Generalized q-Method Relative Pose Estimation for UAVs with Onboard Sensor Measurements
Stanfield, K., Younes, A. B., & Hayajneh, M. (2025) · Sensors · doi:10.3390/s25061939
“Several researchers have focused on enhancing these techniques through the integration of LiDAR and deep learning-based approaches [10] to improve robustness and real-time performance, particularly in GPS-denied environments.”
Beyond Photometric Constraints: Epipolar-Based Self-Supervised Learning for Visual Odometry
Islam, T., & Wang, Y. (2024) · International Conference on Computers and Artificial Intelligence Technology (CAIT) · doi:10.1109/cait64506.2024.10962878
“Wang et al. [27] developed DeepVO, an end-to-end VO model using recurrent CNNs for monocular video.”
SLAM Technology in Autonomous Driving
Xu, J. (2025) · Compendium of Engineering 工程學輯要 · doi:10.63313/engineering.8002
The Integration of Neural Networks in Smart Robotics and Autonomous Systems
Benziane, S. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-4571-0.ch006
Towards End-to-End Visual Odometry for Unstructured Agricultural Environments
Romero-Bautista, V., Altamirano-Robles, L., & Hernández, R. D. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-96255-4_23
LoSeVO: Local Sequence Constraints for Deep Visual Odometry
Zhang, X., Zhu, R., Song, R., He, D., Yang, T., Xiao, Z., & Yan, B. (2025) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas56072.2025.11043333
“Some of the best-known methods are DeepVO [7], GeoNet [8], DeepTAM [9], DeepAVO [10], etc.”
Path Reconstruction from Video Sequences Using Nadiral View and Optical Flow
Ivanovici, M., Plajer, I. C., Popa, Ș., Florea, C., & Racoviţeanu, A. (2025) · International Conference on Engineering of Modern Electric Systems · doi:10.1109/emes65692.2025.11045597
“The advance of machine learning impacted this field too and learning-based VO leverage deep learning models to predict motion directly from image sequences either supervised [8] or unsupervised [9].”
Hanwen, 张. Z., & Yanyang, 王. W. (2025) · Laser & Optoelectronics Progress · doi:10.3788/lop241942
Tracking and Registration Technology Based on Panoramic Cameras
Xu, C., Li, G., Bai, Y., Bai, Y., Cao, Z., & Han, C. (2025) · Applied Sciences · doi:10.3390/app15137397
Robust outdoor trajectory mapping using CNN features and loop closure optimization
Kazi, K., Khizer, A. N., Memon, F., Memon, T. R., & Memon, A. R. (2025) · Mehran University Research Journal of Engineering and Technology · doi:10.22581/muet1982.0351
A Robust and Energy-Efficient Control Policy for Autonomous Vehicles with Auxiliary Tasks
Xu, Y., Yang, C., & Gong, X. (2025) · Electronics · doi:10.3390/electronics14152919
VINS-MultiCam: An ASLfeat-Based MultiCam Visual-Inertial Odometry Framework
Kong, W., Sun, J. W., Wang, C., Dong, W., Chen, X., & Deng, F. Y. (2025) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc65474.2025.11091018
“Moreover, the traditional feature algorithm in frontend is hard to handle the feature tracking in challenging scenarios like image illumination change, motion blur, or large viewpoint change [14].”
Visual Slam in the Heterogeneous Intelligence Era: A SurveyPreprint
Su, S., He, X., Meng, X., Zhang, Y., Yin, W., Mo, L., & Yu, F. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5383002
Spiking Neural Network Based Object Pose Alignment
Yadav, S., Chundawat, C. S., Chaudhary, S., & Kumar, R. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-96-5955-5_33
Enhancing Visual Localization Accuracy in Dynamic Environments via Masking Techniques
Yan, H., Ishihara, A., Shimada, T., Kong, X., & Tomiyama, H. (2025) · International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC) · doi:10.1109/itc-cscc66376.2025.11137722
“Furthermore, with advancements in deep learning, Wang et al. [6] proposed DeepVO, an end-to-end visual odometry method based on deep recurrent convolutional neural networks.”
Shi, W., & He, Y. (2025) · Springer eBooks · doi:10.1007/978-3-031-99485-2_6
Survey on Real-Time Localization and Mapping Algorithms in the Field of Vision
Jiang, X., Cao, J. L., & Li, Z. (2025) · Chinese Control Conference (CCC) · doi:10.23919/ccc64809.2025.11179088
“A notable example is Deep vo [26] , which implements an end-to-end neural network.”
A transformer-enhanced geometric learning framework for robust monocular visual odometry
Pan, Z., Bian, F., Li, X., Hou, J., & Yu, L. (2025) · Transactions of the Institute of Measurement and Control · doi:10.1177/01423312251377711
“In addition, we compared the average inference time on KITTI sequences 00, 02, 05, and 08 with two representative end-to-end methods: DeepVO and TSformer-VO.”
InfluentialVIOSem: Visual-Inertial Odometry via Semantic Communication-enhanced Modulation Design
Wijewardena, K., Klevering, G., Zhang, X., & Xiao, L. (2025) · IEEE International Conference on Mobile Ad-Hoc and Sensor Systems (MASS) · doi:10.1109/mass66014.2025.00035
“Traditional VO systems focus on performing tasks on-device [10], or transmitting image and other integrated sensor data to an edge device for inference [11], [12].”
WLVO: Wavelet Attention Based Self-Supervised Monocular Visual Odometry
Mi, Y., & Shi, H. (2025) · International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) · doi:10.1109/cyberc66434.2025.00047
Robust Underwater Localization via Sensor Fusion and Transformer-Based Visual Odometry
Mughal, M. W., SUn, S., Mehdi, S. A., & Maurelli, F. (2025) · doi:10.23919/oceans59106.2025.11245027
“DeepVO [11] employs recurrent convolutional networks to jointly learn feature representation and camera motion estimation, capturing both spatial and temporal dependencies.”
Xu, F., Almalıoğlu, Y., & Trigoni, N. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11247274
“Supervised methods train end-to-end 6-DoF pose regressors to predict VO [12], [13], and [14].”
Self-citationA Multi-Level Feature Fusion-Based Network for Motion Estimation
Wang, H., Gao, S., Yu, S., Sun, R., & Sun, L. (2025) · IEEE International Conference on Cyborg and Bionic Systems (CBS) · doi:10.1109/cbs65871.2025.11267742
“Approaches like DeepVO[9], UNDeepVO[10], and L-VO[11] treated the problem as a sequence-to-sequence task, directly regressing pose from input images without explicitly handling feature matching, scale, or geometric constraints.”
SF-Loc: A Visual Mapping and Geo-Localization System Based on Sparse Visual Structure Frames
Zhou, Y., Li, X., Li, S., Xia, C., Wang, X., & Feng, S. (2025) · IEEE/ASME Transactions on Mechatronics · doi:10.1109/tmech.2025.3625563
“D Gaussian splatting (3DGS) [19], data-driven association [20], [21], [22], and end-to-end pipelines [23].”
Robust Monocular Visual Odometry via Dual-Paradigm Curriculum Learning
Lahiany, A., & Gal, O. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3635447
“DeepVOmeth-odsevolvedfromsupervisedapproaches(DeepVO[10])to self-supervisedframeworks(UnDeepVO[11],DPVO[12])that learnrobustfeaturesend-to-end,yetstilldriftonhigh-velocity trajectoriesandstrugglewithdomaintransfer.”
Li, D., Yang, T., Wang, Y., Qin, H., Fan, S., Zhang, S., Zhang, Y., & Li, J. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3638876
“Deep learning-based SLAM methods such as DROID-SLAM [21], DeepVO [22], and UnDeepVO [23] have shown promise in improving robustness to challenging visual conditions.”
A Visual SLAM System in Dynamic Environments Based on YOLO11 Instance Segmentation
Wang, X., Li, C., Jing, H., Li, J., Kuang, B., & Jia, G. (2025) · IEEE International Conference on Unmanned Systems (ICUS) · doi:10.1109/icus66297.2025.11295290
“Module replacement integration: A deep learning module is used to replace a key sub-module in traditional SLAM, such as tracking or loop closure (e.g., Deep VO [6]).”
Neural Visual Odometry Based on Improved Detector-Free Matching for Low-Texture Environments
Zhuang, L., Li, W., Meng, X., & Hu, Y. H. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-4875-0_19
6-DoF Object Pose Estimation via Improved Attitude Optimization Network
Wang, Y., An, K., & Zhang, H. (2026) · Lecture notes in electrical engineering · doi:10.1007/978-981-95-4049-5_38
KSLAM: Tightly Coupled Rigid-Body Kinematics for Pose Optimization in Visual SLAM
Dong, T., Kang, K., He, W., Wang, J., & Han, W. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-5740-0_17
Long-Term Feature Point Tracking for Camera Pose Estimation in Forest Fire Scenes
Zhang, H., & Lu, X. (2025) · IEEE International Conference on Big Data and Artificial Intelligence (BDAI) · doi:10.1109/bdai66031.2025.11325611
“DeepVO [12] combines convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to model sequential data, integrating geometric methods with representations, knowledge, and models learned through deep neural networks to enhance VO accuracy.”
An Integrated Smart Road Stud-Based Vehicle Localization Method With Velocity Estimation
Wang, K., Mao, G., Ren, X., Du, H., Huang, B., Fu, T., & Zhang, Z. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3654932
“Specifically, we introduce Speed-Net, a DNN that estimates the ego vehicle’s speed directly from image sequences, built upon an enhanced version of DeepVO [27].”
A Shared-Ground-Target-Groups-Based Cooperative Localization Method for UAVS
Li, X., Ni, C., She, J., & Liu, Y. (2025) · International Conference on Unmanned Systems and Technology (UST) · doi:10.1109/ust66533.2025.11340017
“Wang et al. [8] estimated displacement and direction using sequence image data based on end-to-end visual odometry with deep recurrent convolutional neural networks, which enables UAV positioning in GNSS-denied environments.”
Metadata
- Authors
- Wang, S., Clark, R., Wen, H., & Trigoni, N.
- Affiliations
- University of Oxford (GB)
- Venue
- 2017 IEEE International Conference on Robotics and Automation (ICRA)
- Publisher
- IEEE
- Published
- 2017-05
- Pages
- 2043-2050
- ISSN
- 1050-4729 (ISSN-L)
- OpenAlex
- W2598706937
- Cited by
- 970
- References
- 29
- Citations by year
- 2026: 63 · 2025: 120 · 2024: 104 · 2023: 123 · 2022: 130 · 2021: 131 · 2020: 125 · 2019: 116 · 2018: 52 · 2017: 6
- FWCI
- 143.15
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Proceedings - IEEE International Conference on Robotics and Automation/Proceedings (publishedVersion)
- arXiv (Cornell University) (publishedVersion)
- Warwick Research Archive Portal (University of Warwick) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Engineering and Physical Sciences Research Council (EP/M019918/1)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Physical Sciences › Earth and Planetary Sciences › Geology › 3D Surveying and Cultural Heritage
- Keywords
- monocular visual odometry, end-to-end learning, recurrent convolutional neural network, pose estimation, feature representation learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
