Matches in SemOpenAlex for { <https://semopenalex.org/work/W4328125637> ?p ?o ?g. }
- W4328125637 endingPage "1687" @default.
- W4328125637 startingPage "1687" @default.
- W4328125637 abstract "Object detection in drone-captured images is a popular task in recent years. As drones always navigate at different altitudes, the object scale varies considerably, which burdens the optimization of models. Moreover, high-speed and low-altitude flight cause motion blur on densely packed objects, which leads to great challenges. To solve the two issues mentioned above, based on YOLOv5, we add an additional prediction head to detect tiny-scale objects and replace CNN-based prediction heads with transformer prediction heads (TPH), constructing the TPH-YOLOv5 model. TPH-YOLOv5++ is proposed to significantly reduce the computational cost and improve the detection speed of TPH-YOLOv5. In TPH-YOLOv5++, cross-layer asymmetric transformer (CA-Trans) is designed to replace the additional prediction head while maintain the knowledge of this head. By using a sparse local attention (SLA) module, the asymmetric information between the additional head and other heads can be captured efficiently, enriching the features of other heads. In the VisDrone Challenge 2021, TPH-YOLOv5 won 4th place and achieved well-matched results with the 1st place model (AP 39.43%). Based on the TPH-YOLOv5 and CA-Trans module, TPH-YOLOv5++ can further increase efficiency while achieving comparable and better results." @default.
- W4328125637 created "2023-03-22" @default.
- W4328125637 creator A5029244560 @default.
- W4328125637 creator A5050186100 @default.
- W4328125637 creator A5052052722 @default.
- W4328125637 creator A5069849278 @default.
- W4328125637 creator A5078776547 @default.
- W4328125637 date "2023-03-21" @default.
- W4328125637 modified "2023-10-14" @default.
- W4328125637 title "TPH-YOLOv5++: Boosting Object Detection on Drone-Captured Scenarios with Cross-Layer Asymmetric Transformer" @default.
- W4328125637 cites W1861492603 @default.
- W4328125637 cites W2037227137 @default.
- W4328125637 cites W2565639579 @default.
- W4328125637 cites W2570343428 @default.
- W4328125637 cites W2608264083 @default.
- W4328125637 cites W2797090510 @default.
- W4328125637 cites W2810030371 @default.
- W4328125637 cites W2884585870 @default.
- W4328125637 cites W2962777203 @default.
- W4328125637 cites W2963037989 @default.
- W4328125637 cites W2963351448 @default.
- W4328125637 cites W2963995737 @default.
- W4328125637 cites W2964121718 @default.
- W4328125637 cites W2964241181 @default.
- W4328125637 cites W2966394087 @default.
- W4328125637 cites W2986357608 @default.
- W4328125637 cites W2989604896 @default.
- W4328125637 cites W2989611864 @default.
- W4328125637 cites W2995199175 @default.
- W4328125637 cites W3012573144 @default.
- W4328125637 cites W3034735326 @default.
- W4328125637 cites W3035694605 @default.
- W4328125637 cites W3036271496 @default.
- W4328125637 cites W3046857455 @default.
- W4328125637 cites W3088979451 @default.
- W4328125637 cites W3096609285 @default.
- W4328125637 cites W3106250896 @default.
- W4328125637 cites W3116963012 @default.
- W4328125637 cites W3127743092 @default.
- W4328125637 cites W3138516171 @default.
- W4328125637 cites W3151130473 @default.
- W4328125637 cites W3172087149 @default.
- W4328125637 cites W3176346992 @default.
- W4328125637 cites W3179888767 @default.
- W4328125637 cites W3198011468 @default.
- W4328125637 cites W3205100603 @default.
- W4328125637 cites W3208285567 @default.
- W4328125637 cites W3208871062 @default.
- W4328125637 cites W3210586215 @default.
- W4328125637 cites W3210997132 @default.
- W4328125637 cites W3211328899 @default.
- W4328125637 cites W3211904225 @default.
- W4328125637 cites W4206294875 @default.
- W4328125637 cites W4224132477 @default.
- W4328125637 cites W4226410580 @default.
- W4328125637 cites W4290981008 @default.
- W4328125637 cites W4292184394 @default.
- W4328125637 cites W4312604822 @default.
- W4328125637 cites W4312847199 @default.
- W4328125637 cites W4313007769 @default.
- W4328125637 cites W4313827729 @default.
- W4328125637 cites W4316673310 @default.
- W4328125637 doi "https://doi.org/10.3390/rs15061687" @default.
- W4328125637 hasPublicationYear "2023" @default.
- W4328125637 type Work @default.
- W4328125637 citedByCount "13" @default.
- W4328125637 countsByYear W43281256372023 @default.
- W4328125637 crossrefType "journal-article" @default.
- W4328125637 hasAuthorship W4328125637A5029244560 @default.
- W4328125637 hasAuthorship W4328125637A5050186100 @default.
- W4328125637 hasAuthorship W4328125637A5052052722 @default.
- W4328125637 hasAuthorship W4328125637A5069849278 @default.
- W4328125637 hasAuthorship W4328125637A5078776547 @default.
- W4328125637 hasBestOaLocation W43281256371 @default.
- W4328125637 hasConcept C119599485 @default.
- W4328125637 hasConcept C127413603 @default.
- W4328125637 hasConcept C153180895 @default.
- W4328125637 hasConcept C154945302 @default.
- W4328125637 hasConcept C165801399 @default.
- W4328125637 hasConcept C31972630 @default.
- W4328125637 hasConcept C41008148 @default.
- W4328125637 hasConcept C46686674 @default.
- W4328125637 hasConcept C54355233 @default.
- W4328125637 hasConcept C59519942 @default.
- W4328125637 hasConcept C66322947 @default.
- W4328125637 hasConcept C79403827 @default.
- W4328125637 hasConcept C86803240 @default.
- W4328125637 hasConceptScore W4328125637C119599485 @default.
- W4328125637 hasConceptScore W4328125637C127413603 @default.
- W4328125637 hasConceptScore W4328125637C153180895 @default.
- W4328125637 hasConceptScore W4328125637C154945302 @default.
- W4328125637 hasConceptScore W4328125637C165801399 @default.
- W4328125637 hasConceptScore W4328125637C31972630 @default.
- W4328125637 hasConceptScore W4328125637C41008148 @default.
- W4328125637 hasConceptScore W4328125637C46686674 @default.
- W4328125637 hasConceptScore W4328125637C54355233 @default.
- W4328125637 hasConceptScore W4328125637C59519942 @default.
- W4328125637 hasConceptScore W4328125637C66322947 @default.