Clustering-based Feature Representation Learning for Oracle Bone Inscriptions Detection

Chuntao Li; Honglin Pang; Xinran Fu; Xi Yang; Ye Tao

REVIEW 5 major objections 9 minor 50 references

Clustering-based Feature Representation Learning for Oracle Bone Inscriptions Detection

T0 review · 5 major / 9 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Adding a clustering loss that anchors rubbing-image character features to clean font-library glyphs improves oracle bone inscription detection across Faster R-CNN, DETR, and Sparse R-CNN.

desk verdict A genuinely useful domain trick — font-library-anchored contrastive loss for OBI detection — but the empirical claim is undercut by missing split details and a weak equation. read the letter →

arxiv 2508.18641 v1 pith:UK7TUMT5 submitted 2025-08-26 cs.CV cs.AI

Ye Tao , Xinran Fu , Honglin Pang , Xi Yang , Chuntao Li This is my paper · ORCID

classification cs.CVcs.AI

keywords oracleboneinscriptionsdetectionclustering-basedrepresentationlearningcontrastivefontlibrarypriorknowledgeobjectdegradedarchaeologicalimageryK-Meansclustering

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper tries to establish that a detection network can be made to separate genuine oracle bone characters from cracks, noise, and other look-alike structures by adding a clustering-based contrastive loss that pulls each character-region feature toward a clean, expert-drawn font-library glyph representation and pushes it away from negative cluster centers. The method needs no extra pre-trained network: font-library images and rubbing images pass through the same feature extractor during training, and the extra loss is simply added to the standard classification and box losses. If true, this means curated font libraries can act as reusable prior knowledge for detection in degraded archaeological imagery, improving the accuracy of three mainstream detectors (Faster R-CNN, DETR, Sparse R-CNN) on two oracle bone datasets with modest extra training cost. The paper's evidence is the consistent AP, AP50, and F1-score gains across frameworks and the feature-space visualizations showing character features become linearly separable from non-character features.

What carries the argument

The load-bearing mechanism is the clustering-based contrastive loss $\mathcal{L}_{clus}$ computed on RoI feature vectors. During training, K-Means clusters the negative-anchor features from rubbing images into centers $\{C_N\}$ and clusters positive-anchor features from the OBC font-library images into centers $\{C_M\}$, then uses the average font center $C_M^{\text{mean}}$ as the positive anchor for every sample feature. The loss is a softmax cross-entropy over similarities between the sample feature and the positive center versus all negative centers, with temperature $\tau$ controlling how strongly hard negatives are penalized. The settings of $\tau$, of the loss weight $\lambda_1$, and of the number of font glyphs are model-specific, and the paper shows that too many font anchors (OBC=50) or a different clustering method (DBSCAN) can make the auxiliary loss harmful.

What would settle it

Train the same three detectors with the OBC font-library images replaced by an equal number of clean glyph images from an unrelated script, keeping the clustering loss and all hyperparameters fixed; if average precision still rises, the oracle-font prior is not the active ingredient.

Watch

Extended reading notes

Core claim

The central discovery is that RoI features of oracle bone characters in rubbing images, which initially overlap with non-character structures, can be reorganized by an anchor-based contrastive loss that uses the OBC font library as a clean reference. The method clusters negative-anchor features into $N$ centers $C_N$ and positive font-glyph features into $M$ centers whose average is $C_M^{\text{mean}}$, then applies the loss $\mathcal{L}_{clus}(p_n) = -\log \frac{\exp(p_n \cdot C_M^{\text{mean}}/\tau)}{\sum_n \exp(p_n \cdot C_N/\tau)}$, pulling character features toward the average font-glyph center and away from non-character centers. This loss is added to the standard losses as $\lambda_1 \mathcal{L}_{clus} + \lambda_2 \mathcal{L}_{class} + \lambda_3 \mathcal{L}_{box}$, with the font images contributing only to $\mathcal{L}_{clus}$. The paper reports that this reorganization improves average precision, AP50, and F1-score over the unmodified detectors on both the OBIs detection dataset and OBIMD, and its t-SNE visualizations show that character and non-character features become linearly separable.

Load-bearing premise

The clean, expert-written font-library glyphs must live close enough to the degraded rubbing characters in feature space that pulling the two together separates characters from noise instead of corrupting the detector.

Editorial extensions

If this is right

Any proposal-based detector that produces RoI or query features can take the extra loss without architectural change, so the improvement should transfer to other instance-level detectors beyond the three tested.
Because the font-library prior compacts the character feature region and separates it from crack and noise structures, false positives under heavy degradation should drop.
For sparse-query detectors, smaller temperature values compensate for the smaller pool of negative samples and recover most of the gain.
The extra training cost is bounded and inference cost is unchanged, making the auxiliary loss a cheap add-on for production pipelines.
Using more font glyphs is not monotonically better; the best number of anchors is a tuned hyperparameter, and too many can degrade the model.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The same anchor-cluster recipe could be tried for other degraded writing systems that have clean reference fonts, such as bronze inscriptions or seal script; the authors gesture at this generality but do not test it.
Because the loss collapses all positive font centers into one average point, it discards stylistic variation; a multi-positive or per-cluster contrastive form might preserve more information and could outperform the single-anchor version.
The mechanism predicts that gains will shrink as the input images get cleaner, since the separation the loss enforces has less room to help; this could be checked by adding synthetic clean vs. heavily degraded test splits.
Since the font images contribute only to the auxiliary loss, the idea could be grafted onto character recognition or retrieval tasks, turning a font library into a general feature regularizer.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The core idea is simple and worth taking seriously: use a clean, expert-curated oracle bone font library as a positive anchor and cluster centers of negative features as negatives in a contrastive loss on RoI features. That is a new combination for OBI detection, and it is refreshingly easy to bolt onto standard detectors. The gains in Tables 2 and 3 are small but mostly positive, and the paper does not oversell them in prose. The t-SNE visualizations actually show the intended separation. Good.

The soft spots are mostly about evidence quality. The paper never states how either dataset is split into train/val/test. Hyperparameters — OBC count, temperature, clusters, loss weights — are tuned via Bayesian optimization, and the reported test numbers look like they were selected on the same evaluation set. That makes the +1–3 AP50 claims vulnerable to selection bias. The paper even admits one negative result against a Swin backbone, which is fine, but it undercuts the abstract's “all frameworks demonstrate significant improvements.” The loss equation (Eq. 2) is also inconsistent with the prose: the denominator sums over negative centers only, not all positive centers, and the notation is sloppy. Table 10 reports variance only across K-means initializations, not training seeds or data splits, so reproducibility across runs is not demonstrated. The code and data links are there, but no commit hash or environment — so only principal reproducibility.

On the circularity concern: I agree with your low burden. The font library is external prior knowledge, and cluster centers come from training features, not from test labels. That part is clean. The feature-comparability assumption (do clean font glyphs live in the same feature space as degraded rubbing characters after the shared backbone?) is plausible, and the OBC=50 results suggest it can break, but the method still works when the count is sensible.

Who is this for? People working on OBI detection or similar degraded-script tasks with a font library at hand. It is a modest, honest improvement, not a field-shaping result. The central claim is probably directionally right, but the missing evaluation protocol prevents a clean verdict.

The paper deserves a serious referee. I would send it to review, but require the authors to specify the train/val/test split, report variance across seeds, fix the equation/prose mismatch, and recalibrate the abstract. That is a solid conditional accept after revision.

Referee Report

5 major / 9 minor

Summary. The paper proposes a clustering-based contrastive learning method for detecting Oracle Bone Inscriptions (OBIs) in rubbing images. The method extracts RoI features from three detection frameworks (Faster R-CNN, DETR, Sparse R-CNN), clusters negative features from background regions and positive features from the OBC font library, and adds a contrastive loss L_clus that pulls positive OBI RoI features toward the mean font cluster center and away from negative cluster centers. The total loss is L = λ1 L_clus + λ2 L_class + λ3 L_box. Experiments on two OBI datasets report improvements in AP, AP50, and F1-score over several baselines, with additional ablations on hyperparameters. The paper includes data and code availability links.

Significance. The central idea—using a clean, expert-curated font library as an anchor in feature space—is a reasonable and potentially useful approach for a domain with scarce clean data. If the claimed improvements are real and generalizable, the method would be a simple plug-in module for OBI detection and possibly other ancient-character recognition tasks. Strengths include end-to-end training without auxiliary networks, use of two real datasets, and public code/data links. However, the empirical case is currently weakened by the missing evaluation protocol, the absence of variance measures for baselines, and a contradiction between the abstract and Table 2.

major comments (5)

[Abstract, §3.2, Table 2] The abstract states that 'all frameworks demonstrate significant performance improvements,' but Table 2 shows that Ours(Faster R-CNN-Swin) decreases AP by 0.2 (47.8 vs 48.0), AP50 by 0.2 (86.6 vs 86.8), and AP75 by 0.6 (48.2 vs 48.8) relative to baseline. The text in §3.2 concedes 'only the accuracy rate on Faster R-CNN-Swin has slightly decreased.' This contradiction must be resolved, and the word 'significant' should be removed or justified with statistical tests.
[§2.4, Eq. (2)] The denominator in Eq. (2) sums exp(p_n · C_N / τ) over negative cluster centers C_N, but the surrounding prose states 'the denominator is the similarity of the feature point and all of its positive sample cluster centers.' This mismatch makes the loss definition ambiguous: if the denominator only contains negatives, the loss is not the standard InfoNCE form and its gradient behavior is unclear. Please correct the equation or the description, and clarify the summation index (which should not reuse n).
[§3.2, §3.3, Tables 2, 3, 7] The paper never specifies how the OBIs detection dataset (9,500 images) and OBIMD (10,077 images) are split into training, validation, and test sets. Table 7 reports results obtained after 'Bayesian Optimization technique to tune our hyperparameters,' but it is not stated whether the optimization monitored performance on the test set. If so, the reported gains could be optimistically biased by selection. The authors must describe the split, the tuning protocol, and whether the reported numbers were obtained on a held-out test set never used for model selection.
[Tables 2, 3, 10] Baselines in Tables 2 and 3 are single runs with no error bars or significance tests, while Table 10 reports mean ± std only for the proposed method and only across K-means initializations. Since many reported gains are small (e.g., +0.7 AP for Sparse R-CNN), the reader cannot judge whether the improvements are meaningful. Please report variance or confidence intervals for the baselines as well, or at least provide a paired significance test over multiple training seeds and data splits.
[§3.3, Table 10] The text says 'we conducted three experiments respectively on three models with the same parameters,' but Table 10 reports that experiments were 'repeated 5 times with different random initializations.' These statements are inconsistent. Also, varying only the K-means initialization does not test reproducibility across training seeds, data splits, or other stochastic elements; the claim 'The results are reproducible when rerunning the code' is therefore overstated.

minor comments (9)

[§2.1] Typo: 'alse' should be 'also' in the sentence about OBIMD.
[§3.2] Typo: 'F1-socre' should be 'F1-score' in the introductory sentence of the main results.
[Table 2] The column header 'AR 50' should read 'AR50' for consistency with the text and other tables.
[§2.3] Use the standard spelling 'RoIAlign' instead of 'RoI Align'.
[§2.4] The notation {C_N} and {C_M} is ambiguous: it is not clear whether C_N denotes a single cluster center or the set of centers. Please distinguish between a cluster index and the set.
[§3.2, Table 2] The model named 'DENO-Swin' in Table 2 refers to DINO (per the text); please use consistent naming throughout.
[Tables 8, 9] The merged cells and overlapping values in Tables 8 and 9 are hard to read; please reformat them into separate rows with clear column boundaries.
[§2.4] The sentence 'We'll conduct a contrastive learning of point and centers of clustering' is ungrammatical and should be rewritten.
[§3.1] The evaluation metrics section does not define how F1-Score50 is computed from per-image predictions; please specify whether it is aggregated over all predictions or averaged per image with a confidence threshold.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the contrastive loss is trained with an external font library and evaluated on independent detection metrics.

full rationale

Walked the derivation chain for the clustering-based contrastive loss Lclus (Eq. 2), the cluster-center construction (Eq. 1), the total loss (Eq. 3), and the experimental comparisons (Tables 2 and 3). The positive anchor C_M_mean is computed by K-means on features of OBC font-library images (Section 2.4), which come from an external, expert-curated dataset [4], not from the evaluation set. The negative centers are computed from OBIs training features. The reported detection metrics AP, AP50, and F1-score are computed from predicted boxes against ground-truth boxes, not from the contrastive loss itself. There is no fitted parameter that is renamed as a prediction, and no uniqueness theorem or prior result from the authors is invoked to force the method's design. The only self-citations [8,9] are contextual descriptions of earlier OBI detection methods and do not bear the weight of the central claim. The reviewer concern about possible test-set hyperparameter tuning (Section 3.3, Table 7) is an evaluation-protocol question, not a circularity reduction, so it does not affect this score.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The central claim rests on four tuned hyperparameters (OBC count, temperature, cluster numbers, loss weights) and three domain assumptions about the font library and clustering. None of these are derived from first principles; they are set by experiments or Bayesian search. The method introduces no invented physical entities.

free parameters (4)

Number of OBC font images used as prior knowledge = 10, 20, 50; optimized values 27 (Faster R-CNN), 40 (DETR), 34 (Sparse R-CNN)
A hyperparameter controlling how many font-library images provide positive anchors; Table 5 shows results vary sharply with this choice, and Table 7 reports Bayesian-optimized values.
Temperature parameter tau = 0.01, 0.05, 0.1; optimized values 0.099, 0.069, 0.073
Controls the sharpness of the contrastive loss in Eq. (2); Table 6 shows sensitivity and Table 7 gives optimized values.
Cluster counts M and N = M,N = 63,3 (Faster R-CNN), 41,2 (DETR), 48,4 (Sparse R-CNN)
Numbers of K-Means centers for positive and negative features, determined by Bayesian optimization in Table 7 and not derived from theory.
Loss weights lambda1, lambda2, lambda3 = Faster R-CNN: 1,1,1; DETR: 1,5,1; Sparse R-CNN: 0.1,1,1
Set manually per model in Table 1 so that Lclus has the same order of magnitude as the other losses; the choice affects results as shown in Table 8.

assumptions (3)

domain assumption The OBC font library (AYJGW) images are clean, expert-confirmed representatives of oracle bone characters, so their features are valid positive anchors for rubbing-image character features.
Invoked in Sections 2.1 and 2.4; if font features are not comparable to degraded rubbing features, pulling features together could hurt performance, as seen in Tables 5 and 9.
domain assumption K-Means clustering of RoI features yields stable cluster centers that summarize negative (noise and crack) and positive (font) feature distributions.
Used in Section 2.4, Eq. (1); the paper acknowledges K-Means initialization sensitivity and reports variance in Table 10.
domain assumption Positive anchor boxes in OBI rubbing images correspond to ground-truth character boxes labeled in the dataset, and negative boxes are everything else.
Relied on in Sections 2.2 and 2.3 to construct sample, positive, and negative features; label noise would propagate into the clustering loss.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Clustering-based Feature Representation Learning for Oracle Bone Inscriptions Detection." pith.science (2026). https://pith.science/paper/UK7TUMT5

@misc{pith2026250818641,
  author       = {Pith},
  title        = {Pith review of: Clustering-based Feature Representation Learning for Oracle Bone Inscriptions Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UK7TUMT5}},
  note         = {Machine review of arXiv:2508.18641}
}

read the original abstract

Oracle Bone Inscriptions (OBIs), play a crucial role in understanding ancient Chinese civilization. The automated detection of OBIs from rubbing images represents a fundamental yet challenging task in digital archaeology, primarily due to various degradation factors including noise and cracks that limit the effectiveness of conventional detection networks. To address these challenges, we propose a novel clustering-based feature space representation learning method. Our approach uniquely leverages the Oracle Bones Character (OBC) font library dataset as prior knowledge to enhance feature extraction in the detection network through clustering-based representation learning. The method incorporates a specialized loss function derived from clustering results to optimize feature representation, which is then integrated into the total network loss. We validate the effectiveness of our method by conducting experiments on two OBIs detection dataset using three mainstream detection frameworks: Faster R-CNN, DETR, and Sparse R-CNN. Through extensive experimentation, all frameworks demonstrate significant performance improvements.

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 30 canonical work pages

[1]

In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp

Epshtein, B., Ofek, E., Wexler, Y.: Detecting text in natural scenes with stroke width transform. In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 2963–2970 (2010). IEEE

work page 2010
[2]

In: 2011 International Conference on Document Analysis and Recognition, pp

Lee, J.-J., Lee, P.-H., Lee, S.-W., Yuille, A., Koch, C.: Adaboost for text detection in natural scene. In: 2011 International Conference on Document Analysis and Recognition, pp. 429–434 (2011). IEEE

work page 2011
[3]

Pattern recognition 31(12), 2055–2076 (1998)

Jain, A.K., Yu, B.: Automatic text location in images and video frames. Pattern recognition 31(12), 2055–2076 (1998)

work page 1998
[4]

In: Journal of Physics: Conference Series, vol

Li, B., Dai, Q., Gao, F., Zhu, W., Li, Q., Liu, Y.: Hwobc-a handwriting oracle bone character recognition database. In: Journal of Physics: Conference Series, vol. 1651, p. 012050 (2020). IOP Publishing

work page 2020
[5]

In: 2019 International Conference on Docu- ment Analysis and Recognition (ICDAR), pp

Huang, S., Wang, H., Liu, Y., Shi, X., Jin, L.: Obc306: A large-scale oracle bone character recognition dataset. In: 2019 International Conference on Docu- ment Analysis and Recognition (ICDAR), pp. 681–688 (2019). https://doi.org/ 10.1109/ICDAR.2019.00114

arXiv 2019
[6]

In: Proceedings of the 2020 9th International Conference on Software and Computer Applications, pp

Liu, G., Xing, J., Xiong, J.: Spatial pyramid block for oracle bone inscription detection. In: Proceedings of the 2020 9th International Conference on Software and Computer Applications, pp. 133–140 (2020) 19

work page 2020
[7]

Applied Mathematics-a Journal of Chinese Universities Series B 12, 224–239 (2021)

Liu, G., Chen, S., Xiong, J., Jiao, Q.: An oracle bone inscription detector based on multi-scale gaussian kernels. Applied Mathematics-a Journal of Chinese Universities Series B 12, 224–239 (2021)

work page 2021
[8]

In: Proceedings of the 2024 7th International Conference on Image and Graphics Processing, pp

Fu, X., Zhou, R.: Shape prior fusion for oracle bone inscriptions detection. In: Proceedings of the 2024 7th International Conference on Image and Graphics Processing, pp. 394–401 (2024)

work page 2024

Show all 50 references

[9]

Heritage Science 12(1), 107 (2024)

Fu, X., Zhou, R., Yang, X., Li, C.: Detecting oracle bone inscriptions via pseudo- category labels. Heritage Science 12(1), 107 (2024)

2024
[10]

In: Proceedings of the IEEE International Conference on Computer Vision, pp

Bodla, N., Singh, B., Chellappa, R., Davis, L.S.: Soft-nms–improving object detec- tion with one line of code. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 5561–5569 (2017)

2017
[11]

In: 2017 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), pp

Ning, C., Zhou, H., Song, Y., Tang, J.: Inception single shot multibox detec- tor for object detection. In: 2017 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), pp. 549–554 (2017). https://doi.org/10.1109/ ICMEW.2017.8026312

2017
[12]

In: Proceedings of the AAAI Conference on Artificial Intelligence, vol

Liao, M., Shi, B., Bai, X., Wang, X., Liu, W.: Textboxes: A fast text detector with a single deep neural network. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 31 (2017)

2017
[13]

In: Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14, pp

Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., Berg, A.C.: Ssd: Single shot multibox detector. In: Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14, pp. 21–37 (2016). Springer

2016
[14]

IEEE transactions on image processing 27(8), 3676–3690 (2018)

Liao, M., Shi, B., Bai, X.: Textboxes++: A single-shot oriented scene text detector. IEEE transactions on image processing 27(8), 3676–3690 (2018)

2018
[15]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

Zhou, X., Yao, C., Wen, H., Wang, Y., Zhou, S., He, W., Liang, J.: East: an efficient and accurate scene text detector. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5551–5560 (2017)

2017
[16]

In: Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part VIII 14, pp

Tian, Z., Huang, W., He, T., He, P., Qiao, Y.: Detecting text in natural image with connectionist text proposal network. In: Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part VIII 14, pp. 56–72 (2016). Springer

2016
[17]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

Shi, B., Bai, X., Belongie, S.: Detecting oriented text in natural images by link- ing segments. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2550–2558 (2017)

2017
[18]

In: Proceedings of 20 the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

Zhang, C., Liang, B., Huang, Z., En, M., Han, J., Ding, E., Ding, X.: Look more than once: An accurate detector for text of arbitrary shapes. In: Proceedings of 20 the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10552–10561 (2019)

2019
[19]

arXiv preprint arXiv:1912.09629 3, 15 (2019)

Liu, Y., He, T., Chen, H., Wang, X., Luo, C., Zhang, S., Shen, C., Jin, L.: Explor- ing the capacity of sequential-free box discretization network for omnidirectional scene text detection. arXiv preprint arXiv:1912.09629 3, 15 (2019)

2019 arXiv
[20]

In: Proceedings of the European Conference on Computer Vision (ECCV), pp

Lyu, P., Liao, M., Yao, C., Wu, W., Bai, X.: Mask textspotter: An end-to-end trainable neural network for spotting text with arbitrary shapes. In: Proceedings of the European Conference on Computer Vision (ECCV), pp. 67–83 (2018)

2018
[21]

IEEE transactions on pattern analysis and machine intelligence 43(2), 532–548 (2021)

Liao, M., Lyu, P., He, M., Yao, C., Wu, W., Bai, X.: Mask textspotter: An end- to-end trainable neural network for spotting text with arbitrary shapes. IEEE transactions on pattern analysis and machine intelligence 43(2), 532–548 (2021)

2021
[22]

In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XI 16, pp

Liao, M., Pang, G., Huang, J., Hassner, T., Bai, X.: Mask textspotter v3: Segmentation proposal network for robust scene text spotting. In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XI 16, pp. 706–722 (2020). Springer

2020
[23]

In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

Baek, Y., Lee, B., Han, D., Yun, S., Lee, H.: Character region awareness for text detection. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9365–9374 (2019)

2019
[24]

In: Proceedings of the AAAI Conference on Artificial Intelligence, vol

Liao, M., Wan, Z., Yao, C., Chen, K., Bai, X.: Real-time scene text detection with differentiable binarization. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, pp. 11474–11481 (2020)

2020
[25]

In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

He, K., Fan, H., Wu, Y., Xie, S., Girshick, R.: Momentum contrast for unsu- pervised visual representation learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9729–9738 (2020)

2020
[26]

https://arxiv.org/abs/2003.04297

Chen, X., Fan, H., Girshick, R., He, K.: Improved Baselines with Momentum Contrastive Learning (2020). https://arxiv.org/abs/2003.04297

2020 arXiv
[27]

https://arxiv.org/abs/2104.02057

Chen, X., Xie, S., He, K.: An Empirical Study of Training Self-Supervised Vision Transformers (2021). https://arxiv.org/abs/2104.02057

2021 arXiv
[28]

https://arxiv.org/abs/2002

Chen, T., Kornblith, S., Norouzi, M., Hinton, G.: A Simple Framework for Con- trastive Learning of Visual Representations (2020). https://arxiv.org/abs/2002. 05709

2020
[29]

https://arxiv.org/abs/2006

Chen, T., Kornblith, S., Swersky, K., Norouzi, M., Hinton, G.: Big Self-Supervised Models are Strong Semi-Supervised Learners (2020). https://arxiv.org/abs/2006. 10029 21

2020
[30]

https://arxiv.org/abs/2006.09882

Caron, M., Misra, I., Mairal, J., Goyal, P., Bojanowski, P., Joulin, A.: Unsuper- vised Learning of Visual Features by Contrasting Cluster Assignments (2021). https://arxiv.org/abs/2006.09882

2021 arXiv
[31]

https://arxiv.org/abs/1511.06335

Xie, J., Girshick, R., Farhadi, A.: Unsupervised Deep Embedding for Clustering Analysis (2016). https://arxiv.org/abs/1511.06335

2016 arXiv
[32]

https://arxiv.org/abs/1604.03628

Yang, J., Parikh, D., Batra, D.: Joint Unsupervised Learning of Deep Represen- tations and Image Clusters (2016). https://arxiv.org/abs/1604.03628

2016 arXiv
[33]

In: Proceedings of the European Conference on Computer Vision (ECCV), pp

Caron, M., Bojanowski, P., Joulin, A., Douze, M.: Deep clustering for unsuper- vised learning of visual features. In: Proceedings of the European Conference on Computer Vision (ECCV), pp. 132–149 (2018)

2018
[34]

https://arxiv.org/abs/1905.01278

Caron, M., Bojanowski, P., Mairal, J., Joulin, A.: Unsupervised Pre-Training of Image Features on Non-Curated Data (2019). https://arxiv.org/abs/1905.01278

2019 arXiv
[35]

In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

Feng, T., Wang, W., Wang, X., Yang, Y., Zheng, Q.: Clustering based point cloud representation learning for 3d analysis. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 8283–8294 (2023)

2023
[36]

Information Sciences 689, 121482 (2025)

Daneshfar, F., Saifee, B.S., Soleymanbaigi, S., Aeini, M.: Elastic deep multi-view autoencoder with diversity embedding. Information Sciences 689, 121482 (2025)

2025
[37]

In: Proceedings of the International Conference on Artificial Intelligence, Information Processing and Cloud Computing, pp

Xing, J., Liu, G., Xiong, J.: Oracle bone inscription detection: a survey of oracle bone inscription detection based on deep learning algorithm. In: Proceedings of the International Conference on Artificial Intelligence, Information Processing and Cloud Computing, pp. 1–8 (2019)

2019
[38]

IEEE transactions on pattern analysis and machine intelligence 39(6), 1137–1149 (2016)

Ren, S., He, K., Girshick, R., Sun, J.: Faster r-cnn: Towards real-time object detection with region proposal networks. IEEE transactions on pattern analysis and machine intelligence 39(6), 1137–1149 (2016)

2016
[39]

In: European Conference on Computer Vision, pp

Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., Zagoruyko, S.: End- to-end object detection with transformers. In: European Conference on Computer Vision, pp. 213–229 (2020). Springer

2020
[40]

: Sparse r-cnn: End-to-end object detection with learnable proposals

Sun, P., Zhang, R., Jiang, Y., Kong, T., Xu, C., Zhan, W., Tomizuka, M., Li, L., Yuan, Z., Wang, C., et al. : Sparse r-cnn: End-to-end object detection with learnable proposals. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 14454–14...

2021
[41]

arXiv preprint arXiv:2407.03900 (2024)

Li, B., Luo, D., Liang, Y., Yang, J., Ding, Z., Peng, X., Jiang, B., Han, S., Sui, D., Qin, P., et al.: Oracle bone inscriptions multi-modal dataset. arXiv preprint arXiv:2407.03900 (2024)

2024 arXiv
[42]

https://arxiv.org/abs/1512.03385

He, K., Zhang, X., Ren, S., Sun, J.: Deep Residual Learning for Image Recognition 22 (2015). https://arxiv.org/abs/1512.03385

2015 arXiv
[43]

In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp

Lin, T.-Y., Doll´ ar, P., Girshick, R., He, K., Hariharan, B., Belongie, S.: Feature pyramid networks for object detection. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 936–944 (2017). https://doi.org/ 10.1109/CVPR.2017.106

2017 doi
[44]

https://arxiv.org/abs/1606.04838

Bottou, L., Curtis, F.E., Nocedal, J.: Optimization Methods for Large-Scale Machine Learning (2018). https://arxiv.org/abs/1606.04838

2018 arXiv
[45]

ArXiv abs/1711.05101 (2017)

Loshchilov, I., Hutter, F.: Fixing weight decay regularization in adam. ArXiv abs/1711.05101 (2017)

2017 arXiv
[46]

https://arxiv.org/abs/2010

Powers, D.M.W.: Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation (2020). https://arxiv.org/abs/2010. 16061

2020
[47]

arXiv preprint arXiv:2203.03605 (2022)

Zhang, H., Li, F., Liu, S., Zhang, L., Su, H., Zhu, J., Ni, L.M., Shum, H.-Y.: Dino: Detr with improved denoising anchor boxes for end-to-end object detection. arXiv preprint arXiv:2203.03605 (2022)

2022 arXiv
[48]

In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., Guo, B.: Swin trans- former: Hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 10012–10022 (2021)

2021
[49]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: Unified, real-time object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 779–788 (2016)

2016
[50]

Journal of Machine Learning Research 9(86), 2579–2605 (2008) 23

Maaten, L., Hinton, G.: Visualizing data using t-sne. Journal of Machine Learning Research 9(86), 2579–2605 (2008) 23

2008

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp

Epshtein, B., Ofek, E., Wexler, Y.: Detecting text in natural scenes with stroke width transform. In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 2963–2970 (2010). IEEE

work page 2010

[2] [2]

In: 2011 International Conference on Document Analysis and Recognition, pp

Lee, J.-J., Lee, P.-H., Lee, S.-W., Yuille, A., Koch, C.: Adaboost for text detection in natural scene. In: 2011 International Conference on Document Analysis and Recognition, pp. 429–434 (2011). IEEE

work page 2011

[3] [3]

Pattern recognition 31(12), 2055–2076 (1998)

Jain, A.K., Yu, B.: Automatic text location in images and video frames. Pattern recognition 31(12), 2055–2076 (1998)

work page 1998

[4] [4]

In: Journal of Physics: Conference Series, vol

Li, B., Dai, Q., Gao, F., Zhu, W., Li, Q., Liu, Y.: Hwobc-a handwriting oracle bone character recognition database. In: Journal of Physics: Conference Series, vol. 1651, p. 012050 (2020). IOP Publishing

work page 2020

[5] [5]

In: 2019 International Conference on Docu- ment Analysis and Recognition (ICDAR), pp

Huang, S., Wang, H., Liu, Y., Shi, X., Jin, L.: Obc306: A large-scale oracle bone character recognition dataset. In: 2019 International Conference on Docu- ment Analysis and Recognition (ICDAR), pp. 681–688 (2019). https://doi.org/ 10.1109/ICDAR.2019.00114

arXiv 2019

[6] [6]

In: Proceedings of the 2020 9th International Conference on Software and Computer Applications, pp

Liu, G., Xing, J., Xiong, J.: Spatial pyramid block for oracle bone inscription detection. In: Proceedings of the 2020 9th International Conference on Software and Computer Applications, pp. 133–140 (2020) 19

work page 2020

[7] [7]

Applied Mathematics-a Journal of Chinese Universities Series B 12, 224–239 (2021)

Liu, G., Chen, S., Xiong, J., Jiao, Q.: An oracle bone inscription detector based on multi-scale gaussian kernels. Applied Mathematics-a Journal of Chinese Universities Series B 12, 224–239 (2021)

work page 2021

[8] [8]

In: Proceedings of the 2024 7th International Conference on Image and Graphics Processing, pp

Fu, X., Zhou, R.: Shape prior fusion for oracle bone inscriptions detection. In: Proceedings of the 2024 7th International Conference on Image and Graphics Processing, pp. 394–401 (2024)

work page 2024