The DOI matches its Crossref record; 47 of the 50 references it cites were found in a registry or an index; 3 of the 15 sources its claims cite were checked for support.
Keypoint Detection and Description through Deep Learning in Unstructured Environments
Open accessAbstract
Feature extraction plays a crucial role in computer vision and autonomous navigation, offering valuable information for real-time localization and scene understanding. However, although multiple studies investigate keypoint detection and description algorithms in urban and indoor environments, far fewer studies concentrate in unstructured environments. In this study, a multi-task deep learning architecture is developed for keypoint detection and description, focused on poor-featured unstructured and planetary scenes with low or changing illumination. The proposed architecture was trained and evaluated using a training and benchmark dataset with earthy and planetary scenes. Moreover, the trained model was integrated in a visual SLAM (Simultaneous Localization and Maping) system as a feature extraction module, and tested in two feature-poor unstructured areas. Regarding the results, the proposed architecture provides a mAP (mean Average Precision) in a level of 0.95 in terms of keypoint description, outperforming well-known handcrafted algorithms while the proposed SLAM achieved two times lower RMSE error in a poor-featured area with low illumination, compared with ORB-SLAM2. To the best of the authors’ knowledge, this is the first study that investigates the potential of keypoint detection and description through deep learning in unstructured and planetary environments.
Overview
CoChat Summary
TL;DR
The paper develops HF-net2, a multi-task teacher–student deep learning model for keypoint detection and description in unstructured and planetary scenes, and integrates it into an ORB-SLAM2-based visual SLAM system. It reports that HF-net2 outperforms several handcrafted baselines and that the proposed SLAM achieves lower trajectory error than ORB-SLAM2, especially under low illumination.
Key findings
- For keypoint description, HF-net2 achieved mAP 0.98 under illumination changes and 0.95 under viewpoint changes, with matching scores of 0.74 and 0.63, respectively; this outperformed original HF-net, SIFT, and ORB but remained below teacher SuperPoint (mAP 0.99/0.99).
- For keypoint detection, HF-net2 achieved repeatability 0.74 and mAP 0.71 under illumination changes, and repeatability 0.69 and mAP 0.49 under viewpoint changes; it followed SuperPoint and outperformed SIFT and FAST in the reported comparisons, though Harris had higher viewpoint repeatability (0.77).
- In SLAM experiments, the HF-net2-based SLAM had RMSE 0.12 m vs. ORB-SLAM2’s 0.15 m in a rocky low-illumination square path; RMSE 0.21 m vs. 0.34 m in sandy high illumination; and RMSE 0.24 m vs. 0.50 m in sandy artificial low illumination, with standard deviation 0.06 m vs. 0.12 m.
- The model achieved about 16 ms per-frame inference time and 62.5 FPS on a GPU-enabled machine.
Methods
- HF-net2 is based on HF-net and uses a multi-teacher–student architecture: SuperPoint is the teacher for keypoint detection and local description, NetVLAD is the teacher for global description, and MobileNetV3-large is the shared encoder with separate keypoint, local descriptor, and global descriptor decoders.
- Training used 48,000 FPV/rover-based images from Earth, the Moon, and Mars, including 26,000 Earth RGB images, about 13,000 NASA Curiosity Mars images, and about 9,000 artificial lunar rover images; models were trained for 30,000 iterations with RMSProp at learning rates 0.001–0.00001 using TensorFlow on an NVIDIA GeForce GTX 1080 Ti.
- Evaluation used a benchmark of 120 sequences from Earth, Mars, and the Moon, with 60 illumination-change sequences and 60 viewpoint-change sequences, each containing an original image plus five transformed representations; the SLAM system was based on ORB-SLAM2 with HF-net2 as the feature extractor, tested in rocky and sandy scenes with an Intel RealSense D435 RGB-D camera and RTK GNSS ground truth.
Limitations
- The paper states future work will further improve and fine-tune the architecture by enriching the training and evaluation dataset with more rover-based data, and will optimize the SLAM loop-closing module using only the global descriptor instead of a BoW algorithm, which can fail in completely unknown environments.
- Evident from the text: the SLAM validation is limited to two terrestrial rocky and sandy areas with a small set of paths and illumination conditions, not actual planetary deployment.
- Evident from the text: the benchmark includes artificial lunar images and generated illumination/viewpoint transformations, and HF-net2 generally remains below its teacher SuperPoint in the reported detector and descriptor metrics, except in one qualitative artificial lunar scene.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
3 Valid · 2 Not valid · 7 Not checked · 62 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“However, those architectures are trained with general-purpose datasets such as COCO [41]”
What the sources say
- Supportedcites [41] Lin, T., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., and Zitnick, L. (2014, January 6–12). Microsoft COCO: Common Objects in Context. Proceedings of the ECCV, Zurich, Switzerland. source
The Introduction (p. 2) identifies MS COCO as a dataset of common objects in everyday images. Algorithmic Analysis (p. 11) describes models trained on COCO, supporting its use as a training dataset. The citation identifies COCO as an example, rather than documenting which specific architectures the citing authors mean. (LLM, conf=0.91)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“For the evaluation of the proposed architecture and the comparison with other widely used handcrafted algorithms, a benchmark dataset was designed for unstructured and planetary scenes inspired by HPatches dataset [44], one of the most popular evaluation datasets for keypoint detection and description in general-purpose images.”
What the sources say
- Partly supportedcites [44] Balntas, 2020, H-Patches: A Benchmark and Evaluation of Handcrafted and Learned Local Descriptors, IEEE Trans. Pattern Anal. Mach. Intell. sourceExperimental — accuracy not yet measured
The Benchmark design section (p. 3) identifies HPatches as a benchmark for evaluating local image descriptors, supporting the citation’s identification of the dataset that inspired the citing authors’ benchmark. The Introduction (p. 2) says detector evaluation is a future possibility, not an existing focus; our copy also does not establish that HPatches is among the most popular evaluation datasets. The planetary benchmark is the citing authors’ work, not a result attributed to HPatches. (LLM, conf=0.84)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“the proposed architecture provides a mAP (mean Average Precision) in a level of 0.95 in terms of keypoint description”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“outperforming well-known handcrafted algorithms”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Regarding the Earth, dataset contains 26,000 RGB images, captured from construction sites, mountainous areas, sandy beaches and a quarry from the area of Crete, Greece (Figure 4a).”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
15 more claims
Create a free account to see all 74 claims and every source verdict.
- Probably not a substantial claim. 28
- It may state two separate claims as one. 32
- The quote only partly states this claim. 1
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ platform. 4.2. Evaluation and Results of the Proposed Architecture
“the proposed architecture provides the next most accurate results with matching score at a level of 0.75 and 0.65 and a mAP 0.98 and 0.95”
§ platform. 4.2. Evaluation and Results of the Proposed Architecture
“| HF-net2 (proposed) | 0.74 | 0.98 | 0.63 | 0.95 |”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 4.4. Experiments and Results of the Proposed SLAM
“instead of ORB-SLAM2 accuracy, which is further decreased with RMSE 0.50 m and standard deviation 0.12 m.”
§ 4.4. Experiments and Results of the Proposed SLAM
“| ORB-SLAM2 | 0.50 | 0.17 | 0.85 | 0.21 |”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 4.4. Experiments and Results of the Proposed SLAM
“The proposed SLAM proved its robustness, achieving an RMSE error at a level of 0.25 and a standard deviation at a level of 0.05 in both experiments”
§ 4.4. Experiments and Results of the Proposed SLAM
“Concerning the sandy scene, the proposed SLAM provides significant higher accuracy than ORB-SLAM2 in high illumination with RMSE 0.21 m and standard deviation 0.07 instead of ORB-SLAM2, which provides RMSE 0.34 and a standard deviation of 0.12, respectively.”
Citations
Cites
Distinctive Image Features from Scale-Invariant Keypoints
Lowe, D. G. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000029664.99615.94 · Cited by 55,589
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
MobileNetV2: Inverted Residuals and Linear Bottlenecks
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00474 · Cited by 28,755
Speeded-Up Robust Features (SURF)
Bay, H., Ess, A., Tuytelaars, T., & Van Gool, L. (2008) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2007.09.014 · Cited by 13,557 · cited 1×
“…well-known keypoint detectors including Harris [4], Shi-Tomasi [5], FAST [6], AKAZE [7] and keypoint descriptors such as ORB [8], SIFT [9] and SURF [10] have been widely investigated and used in multiple applications of computer vision and photogrammetry for more than three decades, while…”
A Combined Corner and Edge Detector
Harris, C., & Stephens, M. (1988) · Procedings of the Alvey Vision Conference 1988 · doi:10.5244/c.2.23 · Cited by 12,901 · cited 2×
“The handcrafted and well-known keypoint detectors including Harris [4], Shi-Tomasi [5], FAST [6], AKAZE [7] and keypoint descriptors such as ORB [8], SIFT [9] and SURF [10] have been widely investigated and used in multiple applications of computer vision and photogrammetry for more than three…”
ORB: An efficient alternative to SIFT or SURF
Rublee, E., Rabaud, V., Konolige, K., & Bradski, G. (2011) · 2011 International Conference on Computer Vision · doi:10.1109/iccv.2011.6126544 · Cited by 11,037
TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G. S., Davis, A., Dean, J., Devin, M., et al. (2016) · arXiv · doi:10.48550/arxiv.1603.04467 · Cited by 9,408
Shi, J., & Tomasi (1994) · Proceedings of IEEE Conference on Computer Vision and Pattern Recognition CVPR-94 · doi:10.1109/cvpr.1994.323794 · Cited by 7,724
Shi, W., Caballero, J., Huszar, F., Totz, J., Aitken, A. P., Bishop, R., Rueckert, D., & Wang, Z. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.207 · Cited by 7,570
ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras
Mur-Artal, R., & Tardos, J. D. (2017) · IEEE Transactions on Robotics · doi:10.1109/tro.2017.2705103 · Cited by 6,898 · cited 1×
“The authors use Moon and Mars-analogous terrains for the experimentation of the system while comparing it with state-of-the-art SLAM algorithms including ORB-SLAM2 [28] and VINS-MONO [29], proving higher efficiency, especially in a loop closure.”
VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator
Qin, T., Li, P., & Shen, S. (2018) · IEEE Transactions on Robotics · doi:10.1109/tro.2018.2853729 · Cited by 4,669 · cited 1×
“The authors use Moon and Mars-analogous terrains for the experimentation of the system while comparing it with state-of-the-art SLAM algorithms including ORB-SLAM2 [28] and VINS-MONO [29], proving higher efficiency, especially in a loop closure.”
Machine Learning for High-Speed Corner Detection
Rosten, E., & Drummond, T. (2006) · Lecture Notes in Computer Science · doi:10.1007/11744023_34 · Cited by 4,555
BRIEF: Binary Robust Independent Elementary Features
Calonder, M., Lepetit, V., Strecha, C., & Fua, P. (2010) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-15561-1_56 · Cited by 3,597
Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age
Cadena, C., Carlone, L., Carrillo, H., Latif, Y., Scaramuzza, D., Neira, J., Reid, I., & Leonard, J. J. (2016) · IEEE Transactions on Robotics · doi:10.1109/tro.2016.2624754 · Cited by 3,596
SuperPoint: Self-Supervised Interest Point Detection and Description
DeTone, D., Malisiewicz, T., & Rabinovich, A. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2018.00060 · Cited by 3,355
Past, present, and future of simultaneous localization and mapping: Toward the robust-perception age
Cadena, C., Carlone, L., Carrillo, H., Latif, Y., Scaramuzza, D., Neira, J., Reid, I. D., & Leonard, J. J. (2016) · Zurich Open Repository and Archive (University of Zurich) · doi:10.5167/uzh-138892 · Cited by 3,048
NetVLAD: CNN Architecture for Weakly Supervised Place Recognition
Arandjelovic, R., Gronat, P., Torii, A., Pajdla, T., & Sivic, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.572 · Cited by 2,584
Searching for activation functions
Ramachandran, P., Zoph, B., & Le, Q. V. (2017) · arXiv · doi:10.48550/arxiv.1710.05941 · Cited by 2,335 · cited 2×
“…(1) with the piece-wise alternative of the sigmoid function ((cid:2934) (cid:2878) (cid:2871)) described by the Equation (2). called hard-Swish [37] which combines the Swish ReLU6 ( x + 3 ) 6 by the where σ (x) is the sigmoid function and ReLU6 (Recti fi ed Linear Unit) is a modi fi cation of…”
Fast Explicit Diffusion for Accelerated Features in Nonlinear Scale Spaces
Alcantarilla, P., Nuevo, J., & Bartoli, A. (2013) · Procedings of the British Machine Vision Conference 2013 · doi:10.5244/c.27.13 · Cited by 1,173
From Coarse to Fine: Robust Hierarchical Localization at Large Scale
Sarlin, P.-E., Cadena, C., Siegwart, R., & Dymczyk, M. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.01300 · Cited by 1,044
Elsken, T., Metzen, J. H., & Hutter, F. (2019) · The Springer Series on Challenges in Machine Learning · doi:10.1007/978-3-030-05318-5_3 · Cited by 785
The Lunar Orbiter Laser Altimeter Investigation on the Lunar Reconnaissance Orbiter Mission
Smith, D. E., Zuber, M. T., Jackson, G. B., Cavanaugh, J. F., Neumann, G. A., Riris, H., Sun, X., Zellar, R. S., Coltharp, C., Connelly, J., et al. (2010) · Space Science Reviews · doi:10.1007/s11214-009-9512-y · Cited by 515 · cited 3×
“Concerning the Moon’s dataset, it includes about 9000 artificial rover-based images which which is based on the Lunar Orbiter Laser Altimeter [43] combined with the simulation software Concerning the Moon’s dataset, it includes about 9000 arti fi cial rover-based images which generated and released…”
InfluentialThe TUM VI Benchmark for Evaluating Visual-Inertial Odometry
Schubert, D., Goll, T., Demmel, N., Usenko, V., Stuckler, J., & Cremers, D. (2018) · 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8593419 · Cited by 473
A Comparative Analysis of LiDAR SLAM-Based Indoor Navigation for Autonomous Vehicles
Zou, Q., Sun, Q., Chen, L., Nie, B., & Li, Q. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3063477 · Cited by 321 · cited 1×
“The literature abounds with studies that focus on autonomous driving [12,13], indoor navigation [14], 3D reconstruction [15] or inspection [16] in urban and industrial scenes.”
State‐of‐the‐art technologies for UAV inspections
Jordan, S., Moore, J., Hovet, S., Box, J., Perry, J., Kirsche, K., Lewis, D., & Tse, Z. T. H. (2018) · IET Radar, Sonar & Navigation · doi:10.1049/iet-rsn.2017.0251 · Cited by 260
DXSLAM: A Robust and Efficient Visual SLAM System with Deep Features
Li, D., Shi, X., Long, Q., Liu, S., Yang, W., Wang, F., Wei, Q., & Qiao, F. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9340907 · Cited by 189
Tsintotas, K. A., Bampis, L., & Gasteratos, A. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2022.3175656 · Cited by 175 · cited 1×
“…extraction process is crucial for the visual SLAM (Simultane-ously Localization and Mapping) systems that estimate a robot’s location and contructs a map of the surroundings [2] while being able to detect the loops of the camera trajectory, which aid to further optimize the robot’s navigation [3].”
Inzerillo, L., Di Mino, G., & Roberts, R. (2018) · Automation in Construction · doi:10.1016/j.autcon.2018.10.010 · Cited by 171
iBoW-LCD: An Appearance-Based Loop-Closure Detection Approach Using Incremental Bags of Binary Words
Garcia-Fidalgo, E., & Ortiz, A. (2018) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2018.2849609 · Cited by 167
UcoSLAM: Simultaneous localization and mapping by fusion of keypoints and squared planar markers
Muñoz-Salinas, R., & Medina-Carnicer, R. (2020) · Pattern Recognition · doi:10.1016/j.patcog.2019.107193 · Cited by 163
Local Descriptor for Robust Place Recognition Using LiDAR Intensity
Guo, J., Borges, P. V. K., Park, C., & Gawel, A. (2019) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2019.2893887 · Cited by 162
Visual SLAM for autonomous ground vehicles
Lategahn, H., Geiger, A., & Kitt, B. (2011) · 2011 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2011.5979711 · Cited by 159
A Review of SLAM Techniques and Security in Autonomous Driving
Singandhupe, A., & La, H. M. (2019) · 2019 Third IEEE International Conference on Robotic Computing (IRC) · doi:10.1109/irc.2019.00122 · Cited by 147
HPatches: A benchmark and evaluation of handcrafted and learned local descriptors
Balntas, V., Lenc, K., Vedaldi, A., Tuytelaars, T., Matas, J., & Mikolajczyk, K. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2915233 · Cited by 132
Ongoing Evolution of Visual SLAM from Geometry to Deep Learning: Challenges and Opportunities
Li, R., Wang, S., & Gu, D. (2018) · Cognitive Computation · doi:10.1007/s12559-018-9591-8 · Cited by 85
Feature extraction for underwater visual SLAM
Aulinas, J., Carreras, M., Lladó, X., Salví, J., García, R., Prados, R., & Pétillot, Y. (2011) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceans-spain.2011.6003474 · Cited by 79
The Devon Island Rover Navigation Dataset
Furgale, P., Carle, P., Enright, J., & Barfoot, T. D. (2012) · The International Journal of Robotics Research · doi:10.1177/0278364911433135 · Cited by 65
A Review of Keypoints’ Detection and Feature Description in Image Registration
Liu, C., Xu, J., & Wang, F. (2021) · Scientific Programming · doi:10.1155/2021/8509164 · Cited by 47
Santos, L. C., Aguiar, A., dos Santos, F. N., Valente, A. L. G., & Petry, M. R. (2020) · Robotics · doi:10.3390/robotics9040077 · Cited by 45
Where to Look? Predictive Perception With Applications to Planetary Exploration
Otsu, K., Agha–mohammadi, A., & Paton, M. (2017) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2017.2777526 · Cited by 40
Visual SLAM-Based Robotic Mapping Method for Planetary Construction
Hong, S., Bangunharcana, A., Park, J., Choi, M., & Shin, H. (2021) · Sensors · doi:10.3390/s21227715 · Cited by 39
Illumination invariant feature point matching for high-resolution planetary remote sensing images
Wu, B., Zeng, H., & Hu, H. (2018) · Planetary and Space Science · doi:10.1016/j.pss.2018.01.007 · Cited by 20 · cited 1×
“In [25], an improved version of SIFT for high-resolution remote sensing images from Mars and the Moon is proposed, aiming to reinforce the invariance of SIFT in differences due to illumination.”
Giubilato, R., Gentil, C. L., Vayugundla, M., Schuster, M. J., Vidal‐Calleja, T., & Triebel, R. (2022) · Field Robotics · doi:10.55417/fr.2022053 · Cited by 19
Jung, K., Hitchcox, T., & Forbes, J. R. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10160348 · Cited by 14
A Node-Based Method for SLAM Navigation in Self-Similar Underwater Environments: A Case Study
Trabes, E., & Jordán, M. A. (2017) · Robotics · doi:10.3390/robotics6040029 · Cited by 7
Survey of Visual Feature Extraction Algorithms in a Mars-like Environment
Oelsch, M., Van Opdenbosch, D., & Steinbach, E. (2017) · 2017 IEEE International Symposium on Multimedia (ISM) · doi:10.1109/ism.2017.58 · Cited by 7
A Performance comparison of feature detectors for planetary rover mapping and localization
Wan, W., Peng, M., Xing, Y., Wang, Y., Liu, Z., Di, K., Teng, B., Mao, X., Zhao, Q., Xin, X., et al. (2017) · The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences · doi:10.5194/isprs-archives-xlii-3-w1-149-2017 · Cited by 6 · cited 2×
“On the other hand, study in [23] evaluates the efficiency of the algorithms in terms of several metrics, including repeatability and precision, using simulated images generated with the aid of the DEM (Digital Elevation Model) and DOM (Digital Orthophoto Map) of a Mars region, proving that SIFT…”
Research works printed but not linked: 1
- Howard, A., Sandler, M., Chu, G., Chen, L., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., and Vasudevan, V. (November, January 27). Searching for MobileNetV3. Proceedings of the ICCV, Seoul, Republic of Korea.
Cited by
Zandler, H., Abermann, J., Robson, B. A., et al. (2025) · Frontiers in Remote Sensing · doi:10.3389/frsen.2025.1586933 · Cited by 4
DK-SLAM: Monocular Visual SLAM with Deep Keypoint Learning, Tracking, and Loop Closing
Qu, H., Zhang, L., Mao, J., Tie, J., He, X., Hu, X., Shi, Y., & Chen, C. (2025) · Applied Sciences · doi:10.3390/app15147838 · Cited by 6
Key-Point-Descriptor-Based Image Quality Evaluation in Photogrammetry Workflows
Matuzevičius, D., Urbanavičius, V., Miniotas, D., Mikučionis, Š., Laptik, R., & Ušinskas, A. (2024) · Electronics · doi:10.3390/electronics13112112 · Cited by 3
Beyond Handcrafted Features: A Deep Learning Framework for Optical Flow and SLAM
Kazi, K., Khizer, A. N., Memon, F., Memon, A. R., & Iqbal, M. (2025) · Journal of Imaging · doi:10.3390/jimaging11050155 · Cited by 1
“Our proposed method represents a step forward in the integration of deep learning [6] with SLAM, offering a more reliable solution for autonomous navigation in diverse real-world environments.”
DeepExtract: Neural Networks Based Image Keypoints Extraction for Indirect SLAM Algorithms
Gurram, M., Uttam, P. K., & Ohol, S. S. (2024) · doi:10.1109/iicaiet62352.2024.10730666 · Cited by 1
“To overcome the limitations of conventional feature detectors, recent works [4,6,7,8,9] [3,10,11,12] achieve this by first capturing a video of the environment and then analysing the incoming frames to build the map and estimate the pose of the camera.”
Metadata
- Authors
- Petrakis, G., & Partsinevelos, P.
- Affiliations
- Technical University of Crete (GR)
- Corresponding author
- Partsinevelos, Panagiotis
- Venue
- Robotics
- Publisher
- MDPI AG
- Published
- 2023-09-30
- Volume
- 12
- Issue
- 5
- Pages
- 137
- ISSN
- 2218-6581 (ISSN-L)
- OpenAlex
- W4387234764
- Semantic Scholar
- 1aceb3c5637dcafb7ff0276973a624114cdb4ff1
- Cited by
- 10
- References
- 50
- Citations by year
- 2026: 3 · 2025: 3 · 2024: 4
- FWCI
- 1.21
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Robotics (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- European Commission
- European Social Fund
- 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 Image and Video Retrieval Techniques
- Physical Sciences › Earth and Planetary Sciences › Geology › 3D Surveying and Cultural Heritage
- Keywords
- keypoint detection, keypoint description, feature extraction, unstructured environments, low-light conditions, visual SLAM, multi-task deep learning, autonomous navigation
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-07, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-07 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.