FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3157738407> ?p ?o ?g. }

• W3157738407 abstract "Scene text detection and recognition have been well explored in the past few years. Despite the progress, efficient and accurate end-to-end spotting of arbitrarily-shaped text remains challenging. In this work, we propose an end-to-end text spotting framework, termed PAN++, which can efficiently detect and recognize text of arbitrary shapes in natural scenes. PAN++ is based on the kernel representation that reformulates a text line as a text kernel (central region) surrounded by peripheral pixels. By systematically comparing with existing scene text representations, we show that our kernel representation can not only describe arbitrarily-shaped text but also well distinguish adjacent text. Moreover, as a pixel-based representation, the kernel representation can be predicted by a single fully convolutional network, which is very friendly to real-time applications. Taking the advantages of the kernel representation, we design a series of components as follows: 1) a computationally efficient feature enhancement network composed of stacked Feature Pyramid Enhancement Modules (FPEMs); 2) a lightweight detection head cooperating with Pixel Aggregation (PA); and 3) an efficient attention-based recognition head with Masked RoI. Benefiting from the kernel representation and the tailored components, our method achieves high inference speed while maintaining competitive accuracy. Extensive experiments show the superiority of our method. For example, the proposed PAN++ achieves an end-to-end text spotting F-measure of 64.9 at 29.2 FPS on the Total-Text dataset, which significantly outperforms the previous best method. Code will be available at: https://git.io/PAN." @default.
• W3157738407 created "2021-05-10" @default.
• W3157738407 creator A5006294869 @default.
• W3157738407 creator A5029551526 @default.
• W3157738407 creator A5041031140 @default.
• W3157738407 creator A5045603863 @default.
• W3157738407 creator A5061696740 @default.
• W3157738407 creator A5062687402 @default.
• W3157738407 creator A5068347202 @default.
• W3157738407 creator A5074700770 @default.
• W3157738407 date "2021-05-02" @default.
• W3157738407 modified "2023-09-27" @default.
• W3157738407 title "PAN++: Towards Efficient and Accurate End-to-End Spotting of Arbitrarily-Shaped Text" @default.
• W3157738407 cites W1522301498 @default.
• W3157738407 cites W1686810756 @default.
• W3157738407 cites W1903029394 @default.
• W3157738407 cites W1972065312 @default.
• W3157738407 cites W1988461287 @default.
• W3157738407 cites W2064675550 @default.
• W3157738407 cites W2074849287 @default.
• W3157738407 cites W2108598243 @default.
• W3157738407 cites W2112796928 @default.
• W3157738407 cites W2122221966 @default.
• W3157738407 cites W2127141656 @default.
• W3157738407 cites W2144554289 @default.
• W3157738407 cites W2153182373 @default.
• W3157738407 cites W2194187530 @default.
• W3157738407 cites W2194775991 @default.
• W3157738407 cites W2239285313 @default.
• W3157738407 cites W2253806798 @default.
• W3157738407 cites W2302255633 @default.
• W3157738407 cites W2343052201 @default.
• W3157738407 cites W2519818067 @default.
• W3157738407 cites W2532759528 @default.
• W3157738407 cites W2560023338 @default.
• W3157738407 cites W2565639579 @default.
• W3157738407 cites W2604735854 @default.
• W3157738407 cites W2605076167 @default.
• W3157738407 cites W2605982830 @default.
• W3157738407 cites W2612445135 @default.
• W3157738407 cites W2613718673 @default.
• W3157738407 cites W2740767790 @default.
• W3157738407 cites W2772800855 @default.
• W3157738407 cites W2777652944 @default.
• W3157738407 cites W2784050770 @default.
• W3157738407 cites W2785383245 @default.
• W3157738407 cites W2795619303 @default.
• W3157738407 cites W2810028092 @default.
• W3157738407 cites W2810983211 @default.
• W3157738407 cites W2875814315 @default.
• W3157738407 cites W2896049206 @default.
• W3157738407 cites W2902494497 @default.
• W3157738407 cites W2906333459 @default.
• W3157738407 cites W2949117887 @default.
• W3157738407 cites W2962773189 @default.
• W3157738407 cites W2962790387 @default.
• W3157738407 cites W2962810613 @default.
• W3157738407 cites W2962914239 @default.
• W3157738407 cites W2962986948 @default.
• W3157738407 cites W2963150697 @default.
• W3157738407 cites W2963161243 @default.
• W3157738407 cites W2963299604 @default.
• W3157738407 cites W2963353821 @default.
• W3157738407 cites W2963398399 @default.
• W3157738407 cites W2963403868 @default.
• W3157738407 cites W2963516811 @default.
• W3157738407 cites W2963647456 @default.
• W3157738407 cites W2963712589 @default.
• W3157738407 cites W2963840241 @default.
• W3157738407 cites W2963857746 @default.
• W3157738407 cites W2963977642 @default.
• W3157738407 cites W2964018263 @default.
• W3157738407 cites W2964065044 @default.
• W3157738407 cites W2964294787 @default.
• W3157738407 cites W2964296749 @default.
• W3157738407 cites W2964444661 @default.
• W3157738407 cites W2964685115 @default.
• W3157738407 cites W2965066169 @default.
• W3157738407 cites W2967615747 @default.
• W3157738407 cites W2970910956 @default.
• W3157738407 cites W2981969038 @default.
• W3157738407 cites W2983626510 @default.
• W3157738407 cites W2988098900 @default.
• W3157738407 cites W2991626090 @default.
• W3157738407 cites W3034792612 @default.
• W3157738407 cites W3106228955 @default.
• W3157738407 cites W70975097 @default.
• W3157738407 doi "https://doi.org/10.48550/arxiv.2105.00405" @default.
• W3157738407 hasPublicationYear "2021" @default.
• W3157738407 type Work @default.
• W3157738407 sameAs 3157738407 @default.
• W3157738407 citedByCount "0" @default.
• W3157738407 crossrefType "posted-content" @default.
• W3157738407 hasAuthorship W3157738407A5006294869 @default.
• W3157738407 hasAuthorship W3157738407A5029551526 @default.
• W3157738407 hasAuthorship W3157738407A5041031140 @default.
• W3157738407 hasAuthorship W3157738407A5045603863 @default.
• W3157738407 hasAuthorship W3157738407A5061696740 @default.
• W3157738407 hasAuthorship W3157738407A5062687402 @default.
• W3157738407 hasAuthorship W3157738407A5068347202 @default.

• next