Parkmaster: an in-vehicle, edge-based video analytics service for detecting open parking spaces in urban environments
Open accessThe DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index.
Abstract
We present the design and implementation of ParkMaster, a system that leverages the ubiquitous smartphone to help drivers find parking spaces in the urban environment. ParkMaster estimates parking space availability using video gleaned from drivers' dash-mounted smartphones on the network's edge, uploading analytics about the street to the cloud in real time as participants drive. Novel lightweight parked-car localization algorithms enable the system to estimate each parked car's approximate location by fusing information from phone's camera, GPS, and inertial sensors, tracking and counting parked cars as they move through the driving car's camera frame of view. To visually calibrate the system, ParkMaster relies only on the size of well-known objects in the urban environment for on-the-go calibration. We implement and deploy ParkMaster on Android smartphones, uploading parking analytics to the Azure cloud. On-the-road experiments in three different environments comprising Los Angeles, Paris and an Italian village measure the end-to-end accuracy of the system's parking estimates (close to 90%) as well as the amount of cellular data usage the system requires (less than one mega-byte per hour). Drill-down microbenchmarks then analyze the factors contributing to this end-to-end performance, as video resolution, vision algorithm parameters, and CPU resources.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Histograms of Oriented Gradients for Human Detection
Dalal, N., & Triggs, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.177 · Cited by 33,584 · cited 2×
“In [17, 28, 55, 60] the authors use background subtraction techniques to facilitate object tracking.”
BackgroundMethodA Computational Approach to Edge Detection
Canny, J. (1986) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.1986.4767851 · Cited by 30,004 · cited 1×
“Hence we have investigated edge detection [13] algorithms on the smartphone to detect and measure free parking spaces themselves.”
Method(1996) · Cited by 20,179
Rapid object detection using a boosted cascade of simple features
Viola, P., & Jones, M. (2005) · Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001 · doi:10.1109/cvpr.2001.990517 · Cited by 18,300 · cited 1×
“ParkMaster takes a step further and applies machine learning techniques, the Viola-Jones feature-based cascade classifier [30, 56], to detect complex objects — cars — in the phone camera’s video stream, in real time.”
A flexible new technique for camera calibration
Zhang, Z. (2000) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.888718 · Cited by 15,004 · cited 1×
“, ((uL−1,vL−1) , (XL−1,YL−1,ZL−1))] (3) of (camera, road) coordinate pairs can be passed to OpenCV [39], a computer vision library, which calculates the extrinsic parameters R and t based on a global Levenberg-Marquardt optimization algorithm that minimizes projection error [11, 62].”
BackgroundMean shift: a robust approach toward feature space analysis
Comaniciu, D., & Meer, P. (2002) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.1000236 · Cited by 11,305 · cited 1×
“Preliminary studies have been done in order to choose the proper clustering algorithm: among the approaches which don’t require prior knowledge of the total number of clusters, we evaluated Affinity Propagation [20], Mean Shift [16] and DBSCAN.”
MethodClustering by Passing Messages Between Data Points
Frey, B. J., & Dueck, D. (2007) · Science · doi:10.1126/science.1136800 · Cited by 6,758 · cited 1×
“Preliminary studies have been done in order to choose the proper clustering algorithm: among the approaches which don’t require prior knowledge of the total number of clusters, we evaluated Affinity Propagation [20], Mean Shift [16] and DBSCAN.”
MethodTsai, R. (1987) · IEEE Journal on Robotics and Automation · doi:10.1109/jra.1987.1087109 · Cited by 5,838 · cited 1×
“[53] introduces a pattern-based camera calibration, which has been used by [24] for obstacle avoidance on smartphones (through chessboard camera calibration).”
MethodAn extended set of Haar-like features for rapid object detection
Lienhart, R., & Maydt, J. (2003) · Proceedings. International Conference on Image Processing · doi:10.1109/icip.2002.1038171 · Cited by 2,977 · cited 1×
“ParkMaster takes a step further and applies machine learning techniques, the Viola-Jones feature-based cascade classifier [30, 56], to detect complex objects — cars — in the phone camera’s video stream, in real time.”
MethodA Computational Approach to Edge Detection
CANNY, J. (1987) · Readings in Computer Vision · doi:10.1016/b978-0-08-051581-6.50024-6 · Cited by 2,275
On-road vehicle detection: a review
Sun, Z., Bebis, G., & Miller, R. (2006) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2006.104 · Cited by 1,150 · cited 1×
“Advanced Driver Assistance Systems use computer vision, among other techniques, to alert drivers of potential dangers in the surrounding area [49].”
BackgroundAvidan, S. (2007) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2007.35 · Cited by 1,112 · cited 1×
“[8, 17, 27, 29, 46] apply multi-object tracking-by-detection algorithms, which rely on large temporal video windows to discern among subsequent objects.”
MethodShoup, D. C. (2006) · Transport Policy · doi:10.1016/j.tranpol.2006.05.005 · Cited by 987 · cited 5×
“Focusing on real–time data instead, some cities, such as San Francisco [44], are currently installing sensors on parking meters to detect cars’ presence.”
BackgroundInfluentialPedestrian Detection in Crowded Scenes
Leibe, B., Seemann, E., & Schiele, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.272 · Cited by 858 · cited 1×
“[8, 17, 27, 29, 46] apply multi-object tracking-by-detection algorithms, which rely on large temporal video windows to discern among subsequent objects.”
MethodWu, B., & Nevatia, R. (2007) · International Journal of Computer Vision · doi:10.1007/s11263-006-0027-7 · Cited by 763 · cited 2×
“In [17, 28, 55, 60] the authors use background subtraction techniques to facilitate object tracking.”
BackgroundMethodRobust tracking-by-detection using a detector confidence particle filter
Breitenstein, M. D., Reichlin, F., Leibe, B., Koller-Meier, E., & Van Gool, L. (2009) · 2009 IEEE 12th International Conference on Computer Vision · doi:10.1109/iccv.2009.5459278 · Cited by 545 · cited 2×
“Finally [23, 42] combine particle filters with object detector for Markovian tracking-bydetection, which [12] uses to track pedestrians.”
BackgroundMethodThe University of California: A Pictorial History
Pickerell, A. G., & Dornin, M. (2023) · University of California Press eBooks · doi:10.2307/jj.8501461 · Cited by 515
3D People Tracking with Gaussian Process Dynamical Models
Urtasun, R., Fleet, D. J., & Fua, P. (2006) · 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1 (CVPR'06) · doi:10.1109/cvpr.2006.15 · Cited by 439 · cited 2×
“In [17, 28, 55, 60] the authors use background subtraction techniques to facilitate object tracking.”
BackgroundMethodParkNet: drive-by sensing of road-side parking statistics
Mathur, S., Jin, T., Kasturirangan, N., Chandrasekaran, J., Xue, W., Gruteser, M., & Trappe, W. (2010) · Proceedings of the 8th international conference on Mobile systems, applications, and services · doi:10.1145/1814433.1814448 · Cited by 419 · cited 4×
“The ParkNet System [34] takes a complementary approach, adding ultrasonic sensors on cars to flag empty roadside parking bays.”
BackgroundMethodInfluentialInverse perspective mapping simplifies optical flow computation and obstacle detection
Mallot, H. A., Bülthoff, H. H., Little, J. J., & Bohrer, S. (1991) · Biological Cybernetics · doi:10.1007/bf00201978 · Cited by 392 · cited 1×
“Such techniques however require to know the height of the camera and it’s angle to the road [32], which may change at each run.”
BackgroundSignalGuru: leveraging mobile phones for collaborative traffic signal schedule advisory
Koukoumidis, E., Peh, L.-S., & Martonosi, M. R. (2011) · Proceedings of the 9th international conference on Mobile systems, applications, and services · doi:10.1145/1999995.2000008 · Cited by 316
Measure Locally, Reason Globally: Occlusion-sensitive Articulated Pose Estimation
Sigal, L., & Black, M. J. (2006) · 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2 (CVPR'06) · doi:10.1109/cvpr.2006.180 · Cited by 221 · cited 1×
“[8, 17, 27, 29, 46] apply multi-object tracking-by-detection algorithms, which rely on large temporal video windows to discern among subsequent objects.”
MethodAn image processing system for driver assistance
Handmann, U., Kalinke, T., Tzomakas, C., Werner, M., & Seelen, W. V. (2000) · Image and Vision Computing · doi:10.1016/s0262-8856(99)00032-3 · Cited by 198 · cited 1×
“Several mechanisms have been developed to detect and track cars from a moving vehicle using an in-vehicle camera [9, 22, 38].”
BackgroundCarSafe app: alerting drowsy and distracted drivers using dual cameras on smartphones
You, C.-W., Lane, N. D., Chen, F., Wang, R., Chen, Z., Bao, T. J., Montes-de-Oca, M., Cheng, Y., Lin, M., Torresani, L., et al. (2013) · Proceeding of the 11th annual international conference on Mobile systems, applications, and services · doi:10.1145/2462456.2465428 · Cited by 195
Hierarchical Gaussian process latent variable models
Lawrence, N. D., & Moore, A. J. (2007) · Proceedings of the 24th international conference on Machine learning · doi:10.1145/1273496.1273557 · Cited by 162 · cited 2×
“In [17, 28, 55, 60] the authors use background subtraction techniques to facilitate object tracking.”
BackgroundMethodParkSense: a smartphone based sensing system for on-street parking
Nawaz, S., Efstratiou, C., & Mascolo, C. (2013) · Proceedings of the 19th annual international conference on Mobile computing & networking - MobiCom '13 · doi:10.1145/2500423.2500438 · Cited by 158 · cited 2×
“In contrast with [31, 35, 36, 47], ParkMaster collects data about the surrounding parked cars — not about users actions — thus it doesn’t need to run in every vehicles nor to make additional efforts in guessing non-users’ behavior.”
BackgroundSignalGuru: Leveraging mobile phones for collaborative traffic signal schedule advisory
Koukoumidis, E., Peh, L.-S., & Martonosi, M. (2011) · DSpace@MIT (Massachusetts Institute of Technology) · Cited by 156
Robust Vehicle Detection and Distance Estimation Under Challenging Lighting Conditions
Rezaei, M., Terauchi, M., & Klette, R. (2015) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2015.2421482 · Cited by 153 · cited 1×
“[25, 41, 54] exploit Inverse Perspective Mapping (IPM) to remove perspective effects.”
BackgroundBeyond trees: common-factor models for 2D human pose recovery
Lan, X., & Huttenlocher, D. P. (2005) · Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1 · doi:10.1109/iccv.2005.48 · Cited by 146 · cited 1×
“[8, 17, 27, 29, 46] apply multi-object tracking-by-detection algorithms, which rely on large temporal video windows to discern among subsequent objects.”
MethodMultiple vehicle detection and tracking in hard real-time
Betke, M., Haritaoglu, E., & Davis, L. S. (2002) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.1996.566405 · Cited by 133 · cited 1×
“Several mechanisms have been developed to detect and track cars from a moving vehicle using an in-vehicle camera [9, 22, 38].”
BackgroundToulminet, G., Bertozzi, M., Mousset, S., Bensrhair, A., & Broggi, A. (2006) · IEEE Transactions on Image Processing · doi:10.1109/tip.2006.875174 · Cited by 131 · cited 1×
“Stereo vision is commonly used to detect obstacles and estimate their relative location in the scene using simple triangulation [37, 51, 59].”
BackgroundInvisible Sensing of Vehicle Steering with Smartphones
Chen, D., Cho, K.-T., Han, S., Jin, Z., & Shin, K. G. (2015) · International Conference on Mobile Systems, Applications, and Services · doi:10.1145/2742647.2742659 · Cited by 130 · cited 1×
“While at the moment ParkMaster assumes the user is always doing so, we believe vision-based [40] or sensors-based [6, 7, 10, 15, 43] solutions for lane detection can be easily integrated with ParkMaster to trigger the detection only when the user is driving on the right lane.”
BackgroundHierarchical G aussian Process Latent Variable Models
Lawrence, N. D., & Moore, A. (2007) · Cited by 109
Distance determination for an automobile environment using inverse perspective mapping in OpenCV
Tuohy, S., OCualain, D., Jones, E., & Glavin, M. (2010) · Irish Signals and Systems Conference · doi:10.1049/cp.2010.0495 · Cited by 106 · cited 1×
“[25, 41, 54] exploit Inverse Perspective Mapping (IPM) to remove perspective effects.”
BackgroundHuman tracking with mixtures of trees
Ioffe, S., & Forsyth, D. A. (2002) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2001.937589 · Cited by 106 · cited 1×
“Finally [23, 42] combine particle filters with object detector for Markovian tracking-bydetection, which [12] uses to track pedestrians.”
MethodManweiler, J., Jain, P. K., & Choudhury, R. R. (2012) · International Conference on Mobile Systems, Applications, and Services · doi:10.1145/2307636.2307656 · Cited by 69
SmartLoc: push the limit of the inertial sensor based metropolitan localization using smartphone
Cheng, B., Li, X., Jung, T., Mao, X., Tao, Y., & Yao, L. (2013) · ACM Annual International Conference on Mobile Computing and Networking (MobiCom) · doi:10.1145/2500423.2504574 · Cited by 65 · cited 1×
“While at the moment ParkMaster assumes the user is always doing so, we believe vision-based [40] or sensors-based [6, 7, 10, 15, 43] solutions for lane detection can be easily integrated with ParkMaster to trigger the detection only when the user is driving on the right lane.”
BackgroundNandugudi, A., Ki, T., Nuessle, C., & Challen, G. W. (2014) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/2632048.2632098 · Cited by 62 · cited 3×
“In contrast with [31, 35, 36, 47], ParkMaster collects data about the surrounding parked cars — not about users actions — thus it doesn’t need to run in every vehicles nor to make additional efforts in guessing non-users’ behavior.”
BackgroundLaneQuest: An accurate and energy-efficient lane detection system
Aly, H., Basalamah, A., & Youssef, M. (2015) · IEEE International Conference on Pervasive Computing and Communications · doi:10.1109/percom.2015.7146523 · Cited by 61 · cited 1×
“While at the moment ParkMaster assumes the user is always doing so, we believe vision-based [40] or sensors-based [6, 7, 10, 15, 43] solutions for lane detection can be easily integrated with ParkMaster to trigger the detection only when the user is driving on the right lane.”
BackgroundA high-performance stereo vision system for obstacle detection
Williamson, T. (1998) · Cited by 59
Realtime Depth Estimation and Obstacle Detection from Monocular Video
Wedel, A., Franke, U., Klappstein, J., Brox, T., & Cremers, D. (2006) · Lecture notes in computer science · doi:10.1007/11861898_48 · Cited by 51 · cited 2×
“Furthermore, both [33, 58] require an accurate localization of the camera, which may be hard to achieve with smartphone’s sensors while moving on a car.”
BackgroundAn Algorithm for Parking Lot Occupation Detection
Fabián, T. (2008) · International Conference on Computer Information Systems and Industrial Management Applications (CISIM) · doi:10.1109/cisim.2008.53 · Cited by 50
New Lane Model and Distance Transform for Lane Detection and Tracking
Jiang, R., Klette, R., Vaudrey, T., & Wang, S. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-03767-2_127 · Cited by 50 · cited 1×
“[25, 41, 54] exploit Inverse Perspective Mapping (IPM) to remove perspective effects.”
BackgroundRobust and ubiquitous smartphone-based lane detection
Aly, H., Basalamah, A. M., & Youssef, M. A. (2015) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2015.10.019 · Cited by 45 · cited 1×
“While at the moment ParkMaster assumes the user is always doing so, we believe vision-based [40] or sensors-based [6, 7, 10, 15, 43] solutions for lane detection can be easily integrated with ParkMaster to trigger the detection only when the user is driving on the right lane.”
BackgroundMa, S., Wolfson, O., & Xu, B. R. (2014) · ACM SIGSPATIAL International Workshop on Computational Transportation Science (IWCTS) · doi:10.1145/2674918.2674929 · Cited by 35 · cited 2×
“In contrast with [31, 35, 36, 47], ParkMaster collects data about the surrounding parked cars — not about users actions — thus it doesn’t need to run in every vehicles nor to make additional efforts in guessing non-users’ behavior.”
BackgroundEnhancing in vehicle digital maps via GPS crowdsourcing
Carisi, R., Giordano, E., Pau, G., & Gerla, M. (2011) · Annual Conference on Wireless On-demand Network Systems and Services (WONS) · doi:10.1109/wons.2011.5720196 · Cited by 31 · cited 1×
“The difference is mainly due to higher buildings elevation (especially in the European case) and a more frequent stop-and-go car mobility caused by traffic lights and stops which characterize urban scenarios [14].”
BackgroundSelf-citationCross-Articulation Learning for Robust Detection of Pedestrians
Seemann, E., & Schiele, B. (2006) · Lecture notes in computer science · doi:10.1007/11861898_25 · Cited by 24 · cited 1×
“Finally [23, 42] combine particle filters with object detector for Markovian tracking-bydetection, which [12] uses to track pedestrians.”
MethodRen, F., Huang, J., Terauchi, M., Jiang, R., & Klette, R. (2010) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-642-11577-6_25 · Cited by 24 · cited 1×
“While at the moment ParkMaster assumes the user is always doing so, we believe vision-based [40] or sensors-based [6, 7, 10, 15, 43] solutions for lane detection can be easily integrated with ParkMaster to trigger the detection only when the user is driving on the right lane.”
BackgroundReal-time vehicle tracking and classification
Noll, D., Werner, M., & von Seelen, W. (2002) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.1995.528265 · Cited by 21 · cited 1×
“Several mechanisms have been developed to detect and track cars from a moving vehicle using an in-vehicle camera [9, 22, 38].”
BackgroundToward Real-Time Vehicle Detection Using Stereo Vision and an Evolutionary Algorithm
Nguyen, V. D., Nguyen, T. T., Nguyen, D. D., & Jeon, J. W. (2012) · IEEE Vehicular Technology Conference · doi:10.1109/vetecs.2012.6239921 · Cited by 17 · cited 1×
“Stereo vision is commonly used to detect obstacles and estimate their relative location in the scene using simple triangulation [37, 51, 59].”
BackgroundSekimoto, Y., Matsubayashi, Y., Yamada, H., Ryuichi, I. M. A. I., Usui, T., & Kanasugi, H. (2012) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2012.6338737 · Cited by 15
An efficient obstacle awareness application for Android mobile devices
Itu, R., & Danescu, R. G. (2014) · IEEE International Conference on Intelligent Computer Communication and Processing (ICCP) · doi:10.1109/iccp.2014.6936968 · Cited by 13 · cited 1×
“[53] introduces a pattern-based camera calibration, which has been used by [24] for obstacle avoidance on smartphones (through chessboard camera calibration).”
MethodBluePark: tracking parking and un-parking events in indoor garages
Soubam, S., Banerjee, D., Naik, V. S., & Chakraborty, D. (2016) · International Conference on Distributed Computing and Networking (ICDCN) · doi:10.1145/2833312.2833458 · Cited by 9 · cited 2×
“In contrast with [31, 35, 36, 47], ParkMaster collects data about the surrounding parked cars — not about users actions — thus it doesn’t need to run in every vehicles nor to make additional efforts in guessing non-users’ behavior.”
Background
Research works printed but not linked: 2
- N. True . Vacant parking space detection in static images . University of California , San Diego , 2007 . N. True. Vacant parking space detection in static images. University of California, San Diego, 2007.
- C. Wah . Parking space vacancy monitoring . Projects in Vision and Learning , 2009 . C. Wah. Parking space vacancy monitoring. Projects in Vision and Learning, 2009.
Cited by
All one needs to know about fog computing and related edge computing paradigms: A complete survey
Yousefpour, A., Fung, C., Nguyen, T., Kadiyala, K., Jalali, F., Niakanlahiji, A., Kong, J., & Jue, J. P. (2019) · Journal of Systems Architecture · doi:10.1016/j.sysarc.2019.02.009 · Cited by 1,368
“The authors in [35] design and implement an Android app that helps drivers find parking space availability using MCC.”
BackgroundComputation Offloading Toward Edge Computing
Lin, L., Liao, X., Jin, H., & Li, P. (2019) · Proceedings of the IEEE · doi:10.1109/jproc.2019.2922285 · Cited by 456
“Besides, ParkMaster is another example that uses smartphones as edge nodes to perform data analytics.”
BackgroundKe, R., Zhuang, Y., Pu, Z., & Wang, Y. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2020.2984197 · Cited by 240
“Some other studies explore and test the feasibility of onboard ultrasonic sensors and camera sensors as the moving sensors for crowdsensing [36]–[38].”
BackgroundSmart Parking: Using a Crowd of Taxis to Sense On-Street Parking Space Availability
Bock, F., Martino, S. D., & Origlia, A. (2019) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2019.2899149 · Cited by 165
Video Caching, Analytics, and Delivery at the Wireless Edge: A Survey and Future Directions
Jedari, B., Premsankar, G., Illahi, G. K., Francesco, M. D., Mehrabi, A., & Ylä-Jääski, A. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.3035427 · Cited by 155
Metadata
- Authors
- Grassi, G., Jamieson, K., Bahl, P., & Pau, G.
- Affiliations
- Sorbonne Université (FR)
- University of London (GB)
- Microsoft (United States) (US)
- University of California, Los Angeles (US)
- Venue
- Proceedings of the Second ACM/IEEE Symposium on Edge Computing
- Publisher
- ACM
- Published
- 2017-10-12
- Pages
- 1-14
- OpenAlex
- W2767149267
- Cited by
- 73
- References
- 53
- Citations by year
- 2026: 5 · 2025: 3 · 2024: 11 · 2023: 3 · 2022: 13 · 2021: 7 · 2020: 12 · 2019: 13 · 2018: 4 · 2017: 2
- FWCI
- 13.09
- Type
- conference paper
- Language
- en
- Open access
- Locations
- ACM/IEEE Symposium on Edge Computing (SEC) (publishedVersion)
- HAL (submittedVersion, other-oa)
- Archivio istituzionale della ricerca (Alma Mater Studiorum Università di Bologna) (submittedVersion)
- Funders
- National Science Foundation (1617161)
- Topics
- Social Sciences › Transportation › Smart Parking Systems Research
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Surveillance and Tracking Methods
- Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
- Keywords
- edge computing, multi-sensor fusion, visual calibration, Android
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
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.