Papers › MixNet: Toward Accurate Detection of Challenging Scene Text in the Wild

MixNet: Toward Accurate Detection of Challenging Scene Text in the Wild

23 Aug 2023arXiv:2308.12817archive 2025-07-28

Yu-Xiang Zeng, Jun-Wei Hsieh, Xin Li, Ming-Ching Chang

Detecting small scene text instances in the wild is particularly challenging, where the influence of irregular positions and nonideal lighting often leads to detection errors. We present MixNet, a hybrid architecture that combines the strengths of CNNs and Transformers, capable of accurately detecting small text from challenging natural scenes, regardless of the orientations, styles, and lighting conditions. MixNet incorporates two key modules: (1) the Feature Shuffle Network (FSNet) to serve as the backbone and (2) the Central Transformer Block (CTBlock) to exploit the 1D manifold constraint of the scene text. We first introduce a novel feature shuffling strategy in FSNet to facilitate the exchange of features across multiple scales, generating high-resolution features superior to popular ResNet and HRNet. The FSNet backbone has achieved significant improvements over many existing text detection methods, including PAN, DB, and FAST. Then we design a complementary CTBlock to leverage center line based features similar to the medial axis of text regions and show that it can outperform contour-based approaches in challenging cases when small scene texts appear closely. Extensive experimental results show that MixNet, which mixes FSNet with CTBlock, achieves state-of-the-art results on multiple scene text detection datasets.

PaperPDFCode

Code

D641593/MixNet officialpytorch 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

Scene Text DetectionText Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Scene Text Detection IC19-Art MixNet H-Mean 79.7 #1 of 4 Archive leaderboard report
Scene Text Detection MSRA-TD500 MixNet F-Measure 89.4 #1 of 18 Archive leaderboard report
Scene Text Detection MSRA-TD500 MixNet FPS 15.2 #1 of 18 Archive leaderboard report
Scene Text Detection MSRA-TD500 MixNet Precision 90.7 #1 of 18 Archive leaderboard report
Scene Text Detection MSRA-TD500 MixNet Recall 88.1 #1 of 18 Archive leaderboard report
Scene Text Detection SCUT-CTW1500 MixNet F-Measure 89.8 #1 of 17 Archive leaderboard report
Scene Text Detection SCUT-CTW1500 MixNet FPS 15.2 #1 of 17 Archive leaderboard report
Scene Text Detection SCUT-CTW1500 MixNet Precision 91.4 #1 of 17 Archive leaderboard report
Scene Text Detection SCUT-CTW1500 MixNet Recall 88.3 #1 of 17 Archive leaderboard report
Scene Text Detection Total-Text MixNet F-Measure 90.5% #1 of 27 Archive leaderboard report
Scene Text Detection Total-Text MixNet FPS 15.2 #1 of 27 Archive leaderboard report
Scene Text Detection Total-Text MixNet Precision 93.0 #1 of 27 Archive leaderboard report
Scene Text Detection Total-Text MixNet Recall 88.1 #1 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 ConvolutionAbsolute Position EncodingsAdamAttentionAverage PoolingBPEBatch NormalizationBottleneck Residual BlockConvolutionDense ConnectionsDropoutGlobal Average PoolingGrouped ConvolutionHRNetKaiming InitializationLabel SmoothingLayer NormalizationLinear LayerMax PoolingMixConvMixNetMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUResidual BlockResidual ConnectionSigmoid ActivationSoftmaxSqueeze-and-Excitation BlockTransformer

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