Papers › DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting

DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting

19 Nov 2022CVPR 2023 1arXiv:2211.10772archive 2025-07-28

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

End-to-end text spotting aims to integrate scene text detection and recognition into a unified framework. Dealing with the relationship between the two sub-tasks plays a pivotal role in designing effective spotters. Although Transformer-based methods eliminate the heuristic post-processing, they still suffer from the synergy issue between the sub-tasks and low training efficiency. In this paper, we present DeepSolo, a simple DETR-like baseline that lets a single Decoder with Explicit Points Solo for text detection and recognition simultaneously. Technically, for each text instance, we represent the character sequence as ordered points and model them with learnable explicit point queries. After passing a single decoder, the point queries have encoded requisite text semantics and locations, thus can be further decoded to the center line, boundary, script, and confidence of text via very simple prediction heads in parallel. Besides, we also introduce a text-matching criterion to deliver more accurate supervisory signals, thus enabling more efficient training. Quantitative experiments on public benchmarks demonstrate that DeepSolo outperforms previous state-of-the-art methods and achieves better training efficiency. In addition, DeepSolo is also compatible with line annotations, which require much less annotation cost than polygons. The code is available at https://github.com/ViTAE-Transformer/DeepSolo.

PaperPDFConference PDFCode

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

Code

vitae-transformer/deepsolo 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

DecoderScene Text DetectionText DetectionText MatchingText Spotting

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Text Spotting ICDAR 2015 DeepSolo (ViTAEv2-S, TextOCR) F-measure (%) - Generic Lexicon 79.5 #2 of 18 Archive leaderboard report
Text Spotting ICDAR 2015 DeepSolo (ViTAEv2-S, TextOCR) F-measure (%) - Strong Lexicon 88.1 #2 of 18 Archive leaderboard report
Text Spotting ICDAR 2015 DeepSolo (ViTAEv2-S, TextOCR) F-measure (%) - Weak Lexicon 83.9 #2 of 18 Archive leaderboard report
Text Spotting ICDAR 2015 DeepSolo(ResNet-50, TextOCR) F-measure (%) - Generic Lexicon 79.1 #3 of 18 Archive leaderboard report
Text Spotting ICDAR 2015 DeepSolo(ResNet-50, TextOCR) F-measure (%) - Strong Lexicon 88.0 #3 of 18 Archive leaderboard report
Text Spotting ICDAR 2015 DeepSolo(ResNet-50, TextOCR) F-measure (%) - Weak Lexicon 83.5 #3 of 18 Archive leaderboard report
Text Spotting ICDAR 2015 DeepSolo(ResNet-50) F-measure (%) - Generic Lexicon 76.9 #4 of 18 Archive leaderboard report
Text Spotting ICDAR 2015 DeepSolo(ResNet-50) F-measure (%) - Strong Lexicon 86.8 #4 of 18 Archive leaderboard report
Text Spotting ICDAR 2015 DeepSolo(ResNet-50) F-measure (%) - Weak Lexicon 81.9 #4 of 18 Archive leaderboard report
Text Spotting Total-Text DeepSolo (ViTAEv2-S, TextOCR) F-measure (%) - Full Lexicon 89.6 #1 of 12 Archive leaderboard report
Text Spotting Total-Text DeepSolo (ViTAEv2-S, TextOCR) F-measure (%) - No Lexicon 83.6 #1 of 12 Archive leaderboard report
Text Spotting Total-Text DeepSolo (ResNet-50, TextOCR) F-measure (%) - Full Lexicon 88.7 #2 of 12 Archive leaderboard report
Text Spotting Total-Text DeepSolo (ResNet-50, TextOCR) F-measure (%) - No Lexicon 82.5 #2 of 12 Archive leaderboard report
Text Spotting Total-Text DeepSolo (ResNet-50) F-measure (%) - Full Lexicon 87.0 #3 of 12 Archive leaderboard report
Text Spotting Total-Text DeepSolo (ResNet-50) F-measure (%) - No Lexicon 79.7 #3 of 12 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.

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