Papers › Real-time Scene Text Detection with Differentiable Binarization

Real-time Scene Text Detection with Differentiable Binarization

20 Nov 2019arXiv:1911.08947archive 2025-07-28

Minghui Liao, Zhaoyi Wan, Cong Yao, Kai Chen, Xiang Bai

Recently, segmentation-based methods are quite popular in scene text detection, as the segmentation results can more accurately describe scene text of various shapes such as curve text. However, the post-processing of binarization is essential for segmentation-based detection, which converts probability maps produced by a segmentation method into bounding boxes/regions of text. In this paper, we propose a module named Differentiable Binarization (DB), which can perform the binarization process in a segmentation network. Optimized along with a DB module, a segmentation network can adaptively set the thresholds for binarization, which not only simplifies the post-processing but also enhances the performance of text detection. Based on a simple segmentation network, we validate the performance improvements of DB on five benchmark datasets, which consistently achieves state-of-the-art results, in terms of both detection accuracy and speed. In particular, with a light-weight backbone, the performance improvements by DB are significant so that we can look for an ideal tradeoff between detection accuracy and efficiency. Specifically, with a backbone of ResNet-18, our detector achieves an F-measure of 82.8, running at 62 FPS, on the MSRA-TD500 dataset. Code is available at: https://github.com/MhLiao/DB

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

The archive's official repository (MhLiao/DB) was not harvested; the 25 samples below come from 6 community repositories.

Syntology 25 code samples harvested from 6 repositories linked to this paper: 20 ran (of which 0 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 0 violated, 17 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 25 samples from 6 repositories, 20 ran (17 with no instrument failure; 3 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 3 of the 20 samples that ran have no recorded date.

15 repositories listed; official and paper-mentioned ones first.

MhLiao/DB officialmentioned in papermentioned on GitHubpytorchSyntology: not harvested report
18520339/dbnet-tf2 mentioned on GitHubtfMITSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
2023-MindSpore-1/ms-code-43 mentioned on GitHubmindsporeSyntology: not harvested report
Mushroomcat9998/DBNet mentioned on GitHubpytorchApache-2.0Syntology: 3 samples, 2 ran (0 instrument-failed) · 1 unverified report
PaddlePaddle/PaddleOCR mentioned on GitHubpaddleApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to KRETA: A Benchmark for Korean Reading and Reasoning in Text-Rich VQA Attuned to... (arXiv:2508.19944; that paper's own run record: 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 3 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
SURFZJY/Real-time-Text-Detection mentioned on GitHubpytorchApache-2.0Syntology: 8 samples, 6 ran (2 instrument-failed) · 2 unverified report
WenmuZhou/DBNet.pytorch mentioned on GitHubpytorchApache-2.0Syntology: 3 samples, 1 ran (1 instrument-failed) · 2 unverified report
WenmuZhou/PytorchOCR mentioned on GitHubpytorchSyntology: not harvested report
huyhoang17/DB_text_minimal mentioned on GitHubpytorchMITSyntology: 7 samples, 7 ran (0 instrument-failed) · 0 unverified report
jakeywu/ocr_torch mentioned on GitHubpytorchSyntology: not harvested report
mindee/doctr mentioned on GitHubpytorchApache-2.0Syntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
xuannianz/DifferentiableBinarization mentioned on GitHubtfSyntology: not harvested report
yanan0122/dbnet-and-dbnet_pp-by-mind-spore mentioned on GitHubmindsporeSyntology: not harvested report
mindspore-lab/mindocr mindsporenot reachable when probed 2026-09-17 — repositories for recent papers often appear after camera-readySyntology: not harvested report
open-mmlab/mmocr pytorchApache-2.0Syntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

25 samples harvested: 20 ran (of which 0 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 0 violated, 17 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · our draft was wrong
17ran
5unverified

Licence: 1 of the 25 samples is pointer only, meaning Syntology does not serve that copy's text. For that sample, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 6 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

18520339/dbnet-tf2 community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
crop_patch in 10 papers tfdbnet/utils.py ran MIT (permissive) · 380431820253ae48 · report fixture: signature page:ndarray/3/int, box_points:ndarray/2/float · returned ndarray · level recorded 2026-09-24
resize_image_short_side tfdbnet/utils.py ran fingerprinted MIT (permissive) · a5ebf7edacfab1eb · report fixture: signature image:ndarray/3/int, image_short_side:int/0/int · returned ndarray · level recorded 2026-09-24
Mushroomcat9998/DBNet community (archive-listed) 3 samples, 2 ran (0 instrument-failed) · 1 unverified
resize_image in 74 papers data_loader/modules/augment.py ran Apache-2.0 (permissive) · 74c3e5afa7495d81 · report fixture: signature img:ndarray/3/int, short_size:int/0/int · returned tuple · level recorded 2026-09-24
shrink_polygon_py data_loader/modules/make_shrink_map.py ran fingerprinted Apache-2.0 (permissive) · a48b1586c42ff5fe · report fixture: signature polygon:ndarray/2/float, shrink_ratio:float/0/float · returned ndarray · level recorded 2026-09-24
shrink_polygon_pyclipper data_loader/modules/make_shrink_map.py unverified Apache-2.0 (permissive) · e5fc5488f52a4d68 · report
SURFZJY/Real-time-Text-Detection community (archive-listed) 8 samples, 6 ran (2 instrument-failed) · 2 unverified
conv1x1 in 691 papers models/modules/resnet.py ran · our draft was wrong Apache-2.0 (permissive) · d9def42110729a85 · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int · returned Conv2d · level recorded, date not recorded
conv3x3 in 1,376 papers models/modules/resnet.py ran · our draft was wrong Apache-2.0 (permissive) · 160bb14bd76201b4 · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int, groups:int/0/int, dilation:int/0/int · returned Conv2d · level recorded, date not recorded
get_images in 21 papers genText.py ran fingerprinted Apache-2.0 (permissive) · b1827bd3314d5a57 · report fixture: signature img_path:str/0/str · returned list · level recorded 2026-09-24
get_txts genText.py ran fingerprinted Apache-2.0 (permissive) · acbee53c5f8e1a9a · report fixture: signature txt_path:str/0/str · returned list · level recorded 2026-09-24
quadratic data_loader/data_utils.py ran fingerprinted Apache-2.0 (permissive) · a133283a2aa3440f · report fixture: signature a:int/0/int, b:int/0/int, c:int/0/int · returned tuple · level recorded 2026-09-24
unshrink_offset data_loader/data_utils.py ran Apache-2.0 (permissive) · d1012dc3a8f720ea · report fixture: signature poly:ndarray/3/float, ratio:float/0/float · returned tuple · level recorded 2026-09-24
check_and_validate_polys data_loader/data_utils.py unverified Apache-2.0 (permissive) · 07471b3162a18b9d · report
dcn_conv3x3 models/modules/resnet.py unverified Apache-2.0 (permissive) · f9244a258a55a5ab · report
WenmuZhou/DBNet.pytorch community (archive-listed) 3 samples, 1 ran (1 instrument-failed) · 2 unverified
conv3x3 in 1,376 papers models/backbone/resnet.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · fac5364e2f53c6db · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int · returned Conv2d · level recorded, date not recorded
deformable_resnet18 models/backbone/resnet.py unverified Apache-2.0 (permissive) · ae0dd302eac3e5eb · report
resnet18 in 432 papers models/backbone/resnet.py unverified Apache-2.0 (permissive) · 7bee880a6603be77 · report
huyhoang17/DB_text_minimal community (archive-listed) 7 samples, 7 ran (0 instrument-failed) · 0 unverified
cal_text_score src/text_metrics.py ran MIT (permissive) · 0b16bed946565da6 · report fixture: signature texts:Tensor/3/float, gt_texts:Tensor/3/float, training_masks:Tensor/3/float, running_metric_text:MinimalMetric/0/?, thresh:float/0/float · returned float64 · level recorded 2026-09-24
compute_distance in 21 papers src/db_transforms.py ran MIT (permissive) · 0362f3cc6b5edc59 · report fixture: signature xs:ndarray/1/float, ys:ndarray/1/float, point_1:tuple/1/float, point_2:tuple/1/float · returned ndarray · level recorded 2026-09-24
split_regions src/db_transforms.py ran MIT (permissive) · af7b1b462f45fefc · report fixture: signature axis:ndarray/1/int · returned list · level recorded 2026-09-24
step_function src/losses.py ran MIT (permissive) · 06d815aba9fc17d6 · report fixture: signature x:list/1/float, y:list/1/float, k:int/0/int · returned Tensor · level recorded 2026-09-24
test_preprocess src/db_handler.py ran MIT (permissive) · bcb66632e553eda2 · report fixture: signature img:ndarray/3/int, mean:list/1/float, to_tensor:bool/0/bool, pad:bool/0/bool · returned Tensor · level recorded 2026-09-24
test_resize src/db_handler.py ran MIT (permissive) · 80b87673405e6943 · report fixture: signature img:ndarray/3/int, size:int/0/int, pad:bool/0/bool · returned ndarray · level recorded 2026-09-24
to_list_tuples src/make_eval.py ran MIT (permissive) · a8adaa21874014ed · report fixture: signature sample:list/1/list · returned list · level recorded 2026-09-24
mindee/doctr community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
get_max_width_length_ratio doctr/models/_utils.py ran Apache-2.0 (permissive) · 14c65cfb11d94847 · report fixture: signature contour:ndarray/3/float · returned float · level recorded 2026-09-24
rectify_crops doctr/models/_utils.py ran Apache-2.0 (permissive) · 0ef99817b8bb3cdb · report fixture: signature crops:list/1/ndarray, orientations:list/1/int · returned list · level recorded 2026-09-24

Tasks

BinarizationOptical Character Recognition (OCR)Scene Text DetectionSegmentationText Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Scene Text Detection ICDAR 2015 DB-ResNet-50 (1152) F-Measure 87.3 #13 of 43 Archive leaderboard report
Scene Text Detection ICDAR 2015 DB-ResNet-50 (1152) Precision 91.8 #13 of 43 Archive leaderboard report
Scene Text Detection ICDAR 2015 DB-ResNet-50 (1152) Recall 83.2 #13 of 43 Archive leaderboard report
Scene Text Detection MSRA-TD500 DB-ResNet-50 (736) F-Measure 84.9 #6 of 18 Archive leaderboard report
Scene Text Detection MSRA-TD500 DB-ResNet-50 (736) Precision 91.5 #6 of 18 Archive leaderboard report
Scene Text Detection MSRA-TD500 DB-ResNet-50 (736) Recall 79.2 #6 of 18 Archive leaderboard report
Scene Text Detection SCUT-CTW1500 DB-ResNet50 (1024) F-Measure 83.4 #10 of 17 Archive leaderboard report
Scene Text Detection Total-Text DB-ResNet-50 (800) F-Measure 84.7% #15 of 27 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

1x1 ConvolutionAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionGlobal Average PoolingKaiming InitializationMax PoolingReLUResidual BlockResidual Connection

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections