Confidence Estimation for Object Detection in Document Images
Open accessAbstract
Deep neural networks are becoming increasingly powerful and large and require always more labelled data to be trained. However, since annotating data is time consuming, it is now necessary to develop systems that show good performance while learning on a limited amount of data. These data must be correctly chosen in order to obtain models that are still efficient. For this, the systems must be able to determine which data should be annotated in order to achieve the best results. In this paper, we propose three estimators to estimate the confidence of object detection predictions. The first two are based on Monte Carlo dropout and the third one on descriptive statistics. In active learning framework, the three estimators show a significant improvement in performance for the detection of document physical pages and text lines compared to a random selection of images. We also show that the proposed estimator based on descriptive statistics can replace MC dropout, reducing the computation cost without compromising the performances.
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
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?
Other applications such as document layout detection, pose estimation, and surface defect detection remain under-explored.
“Other applications such as document layout detection [14,17], pose estimation [7], and surface defect detection [55] remain under-explored, as well.”
From Ten Years of Active Learning Techniques and Object Detection: A Systematic Review (2023) · cites this paper as [17]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
The pascal visual object classes (VOC) challenge
Everingham, M., Van Gool, L., Williams, C. K. I., Winn, J., & Zisserman, A. (2010) · International Journal of Computer Vision · doi:10.1007/s11263-009-0275-4 · Cited by 19,599 · see citing works
Proceedings of the 24th international conference on Machine learning
(2007) · doi:10.1145/1273496 · Cited by 11,869 · see citing works
Dropout as a bayesian approximation: Representing model uncertainty in deep learning
Gal, Y., & Ghahramani, Z. (2015) · arXiv · doi:10.48550/arxiv.1506.02142 · Cited by 6,471 · see citing works
Deep neural networks are easily fooled: High confidence predictions for unrecognizable images
Nguyen, A., Yosinski, J., & Clune, J. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298640 · Cited by 3,258 · see citing works
All of Statistics: A Concise Course in Statistical Inference
Wasserman, L. (2004) · Springer Texts in Statistics · doi:10.1007/978-0-387-21736-9 · Cited by 1,090 · see citing works
An analysis of active learning strategies for sequence labeling tasks
Settles, B., & Craven, M. W. (2008) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/1613715.1613855 · Cited by 991 · see citing works
Confidence-Aware Learning for Deep Neural NetworksPreprint
Hwang, S., Moon, J., Kim, J., & Shin, Y. (2020) · International Conference on Machine Learning · doi:10.48550/arxiv.2007.01458 · Cited by 206 · see citing works
Active Learning with Statistical ModelsPreprint
Cohn, D. A., Ghahramani, Z., & Jordan, M. I. (1996) · arXiv (Cornell University) · doi:10.48550/arxiv.cs/9603104 · Cited by 155 · see citing works
READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents
Grüning, T., Labahn, R., Diem, M., Kleber, F., & Fiel, S. (2018) · IAPR International Workshop on Document Analysis Systems (DAS) · doi:10.1109/das.2018.38 · Cited by 86 · see citing works
Active Learning for Deep Object Detection
Brust, C.-A., Käding, C., & Denzler, J. (2019) · Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications · doi:10.5220/0007248601810190 · Cited by 78 · see citing works
Bayesian U-Net: Estimating Uncertainty in Semantic Segmentation of Earth Observation Images
Dechesne, C., Lassalle, P., & Lefèvre, S. (2021) · Remote Sensing · doi:10.3390/rs13193836 · Cited by 56 · see citing works
Active Learning for Deep Object Detection
Brust, C.-A., Käding, C., & Denzler, J. (2019) · Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications · doi:10.5220/0007248600002108 · Cited by 48 · see citing works
Multiple Document Datasets Pre-training Improves Text Line Detection With Deep Neural Networks
Boillet, M., Kermorvant, C., & Paquet, T. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412447 · Cited by 46 · see citing works
Self-citationMeasuring the accuracy of page-reading systems
Rice, S. V. (2019) · Digital Scholarship - UNLV (University of Nevada Reno) · doi:10.25669/hfa8-0cqv · Cited by 45 · see citing works
Robust text line detection in historical documents: learning and evaluation methods
Boillet, M., Kermorvant, C., & Paquet, T. (2022) · International Journal on Document Analysis and Recognition (IJDAR) · doi:10.1007/s10032-022-00395-7 · Cited by 38 · see citing works
Self-citationBoillet, M., Bonhomme, M.-L., Stutzmann, D., & Kermorvant, C. (2019) · International Workshop on Historical Document Imaging and Processing (HIP) · doi:10.1145/3352631.3352633 · Cited by 33 · see citing works
Self-citationEuropeana Newspapers OCR Workflow Evaluation
Pletschacher, S., Clausner, C., & Antonacopoulos, A. (2015) · doi:10.1145/2809544.2809554 · Cited by 25 · see citing works
Arora, A., Garcia, L. P., Watanabe, S., Manohar, V., Shao, Y., Khudanpur, S. P., Chang, C.-C., Rekabdar, B., BabaAli, B., Povey, D., et al. (2019) · IAPR International Conference on Document Analysis and Recognition (ICDAR) · doi:10.1109/icdar.2019.00111 · Cited by 24 · see citing works
A Comprehensive Comparison of Open-Source Libraries for Handwritten Text Recognition in Norwegian
Maarand, M., Beyer, Y., Kåsen, A., Fosseide, K. T., & Kermorvant, C. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-06555-2_27 · Cited by 21 · see citing works
Self-citationReducing the Human Effort in Text Line Segmentation for Historical Documents
Granell, E., Quirós, L., Romero, V., & Sánchez, J. A. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-86334-0_34 · Cited by 8 · see citing works
(2000) · CrossRef Listing of Deleted DOIs · doi:10.1162/153244302760185243
Flagged first, then by citations
Research works printed but not linked: 1
- C. Tensmeyer, B. Davis, C. Wigington, I. Lee, B. Barrett, Pa- geNet: Page Boundary Extraction in Historical Handwritten Documents, in: Proceedings of the 4th International Workshop on Historical Document Imaging and Processing, 2017, pp. 59- 64. doi:10.1145/3151509.3151522. · DOI
Cited by
Sorted by citations
Ten Years of Active Learning Techniques and Object Detection: A Systematic Review
Garcia, D., Carias, J., Adão, T., Jesus, R., Cunha, A., & Magalhães, L. G. (2023) · Applied Sciences · doi:10.3390/app131910667 · Cited by 17 · see citing works
Détection d'objets dans les documents numérisés par réseaux de neurones profonds
Boillet, M. (2023) · theses.fr (ABES) · doi:10.70675/2efdd80ez5832z43feza486za044a9e49b08
Self-citation
- Authors
- Boillet, M., Kermorvant, C., & Paquet, T.
- Affiliations
- Laboratoire d'Informatique, du Traitement de l'Information et des Systèmes (FR)
- Corresponding author
- Boillet, Mélodie
- Venue
- SSRN Electronic Journal
- Publisher
- Elsevier BV
- Published
- 2022
- ISSN
- 1556-5068 (ISSN-L)
- OpenAlex
- W4285308934
- Cited by
- 2
- References
- 17
- Citations by year
- 2023: 2
- Type
- posted content
- Language
- en
- Open access
- green
- doi.org
- Locations
- SSRN Electronic Journal (acceptedVersion)
- Funders
- Joint Programming Initiative on Cultural Heritage
- Agence Nationale de la Recherche (ANR-17-JPCH-0006, ANR-17)
- Association Nationale de la Recherche et de la Technologie
- Norges Forskningsråd
- Grand Équipement National De Calcul Intensif
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Handwritten Text Recognition Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Algorithms
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- confidence estimation, object detection, scanned documents, active learning (machine learning), Monte Carlo dropout, descriptive statistics, data annotation, text line detection, limited labeled data
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
