Papers › DPText-DETR: Towards Better Scene Text Detection with Dynamic Points in Transformer

DPText-DETR: Towards Better Scene Text Detection with Dynamic Points in Transformer

10 Jul 2022arXiv:2207.04491archive 2025-07-28

Maoyuan Ye, Jing Zhang, Shanshan Zhao, Juhua Liu, Bo Du, DaCheng Tao

Recently, Transformer-based methods, which predict polygon points or Bezier curve control points for localizing texts, are popular in scene text detection. However, these methods built upon detection transformer framework might achieve sub-optimal training efficiency and performance due to coarse positional query modeling.In addition, the point label form exploited in previous works implies the reading order of humans, which impedes the detection robustness from our observation. To address these challenges, this paper proposes a concise Dynamic Point Text DEtection TRansformer network, termed DPText-DETR. In detail, DPText-DETR directly leverages explicit point coordinates to generate position queries and dynamically updates them in a progressive way. Moreover, to improve the spatial inductive bias of non-local self-attention in Transformer, we present an Enhanced Factorized Self-Attention module which provides point queries within each instance with circular shape guidance. Furthermore, we design a simple yet effective positional label form to tackle the side effect of the previous form. To further evaluate the impact of different label forms on the detection robustness in real-world scenario, we establish an Inverse-Text test set containing 500 manually labeled images. Extensive experiments prove the high training efficiency, robustness, and state-of-the-art performance of our method on popular benchmarks. The code and the Inverse-Text test set are available at https://github.com/ymy-k/DPText-DETR.

PaperPDFCode

In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.

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

Code

ymy-k/dptext-detr officialmentioned in papermentioned on GitHubpytorchNOASSERTION 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.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

FormInductive BiasScene Text DetectionText Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Scene Text Detection IC19-Art DPText-DETR (ResNet-50) H-Mean 78.1 #4 of 4 Archive leaderboard report
Scene Text Detection SCUT-CTW1500 DPText-DETR (ResNet50) F-Measure 88.8 #3 of 17 Archive leaderboard report
Scene Text Detection SCUT-CTW1500 DPText-DETR (ResNet50) Precision 91.7 #3 of 17 Archive leaderboard report
Scene Text Detection SCUT-CTW1500 DPText-DETR (ResNet50) Recall 86.2 #3 of 17 Archive leaderboard report
Scene Text Detection Total-Text DPText-DETR (ResNet-50) F-Measure 89.0% #3 of 27 Archive leaderboard report
Scene Text Detection Total-Text DPText-DETR (ResNet-50) FPS 17 #3 of 27 Archive leaderboard report
Scene Text Detection Total-Text DPText-DETR (ResNet-50) Precision 91.8% #3 of 27 Archive leaderboard report
Scene Text Detection Total-Text DPText-DETR (ResNet-50) Recall 86.4% #3 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

Deformable Attention ModuleDeformable DETRDetrTestTransformer

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