Matches in SemOpenAlex for { <https://semopenalex.org/work/W3137992142> ?p ?o ?g. }
- W3137992142 abstract "This paper revisits feature pyramids networks (FPN) for one-stage detectors and points out that the success of FPN is due to its divide-and-conquer solution to the optimization problem in object detection rather than multi-scale feature fusion. From the perspective of optimization, we introduce an alternative way to address the problem instead of adopting the complex feature pyramids - {em utilizing only one-level feature for detection}. Based on the simple and efficient solution, we present You Only Look One-level Feature (YOLOF). In our method, two key components, Dilated Encoder and Uniform Matching, are proposed and bring considerable improvements. Extensive experiments on the COCO benchmark prove the effectiveness of the proposed model. Our YOLOF achieves comparable results with its feature pyramids counterpart RetinaNet while being $2.5times$ faster. Without transformer layers, YOLOF can match the performance of DETR in a single-level feature manner with $7times$ less training epochs. With an image size of $608times608$, YOLOF achieves 44.3 mAP running at 60 fps on 2080Ti, which is $13%$ faster than YOLOv4. Code is available at url{https://github.com/megvii-model/YOLOF}." @default.
- W3137992142 created "2021-03-29" @default.
- W3137992142 creator A5000727470 @default.
- W3137992142 creator A5027961969 @default.
- W3137992142 creator A5061145810 @default.
- W3137992142 creator A5073212256 @default.
- W3137992142 creator A5080470862 @default.
- W3137992142 creator A5091093335 @default.
- W3137992142 date "2021-03-17" @default.
- W3137992142 modified "2023-10-11" @default.
- W3137992142 title "You Only Look One-level Feature" @default.
- W3137992142 cites W1536680647 @default.
- W3137992142 cites W1665214252 @default.
- W3137992142 cites W1861492603 @default.
- W3137992142 cites W1901129140 @default.
- W3137992142 cites W1921523184 @default.
- W3137992142 cites W2102605133 @default.
- W3137992142 cites W2168356304 @default.
- W3137992142 cites W2194775991 @default.
- W3137992142 cites W2222512263 @default.
- W3137992142 cites W2407521645 @default.
- W3137992142 cites W2549139847 @default.
- W3137992142 cites W2565639579 @default.
- W3137992142 cites W2570343428 @default.
- W3137992142 cites W2613718673 @default.
- W3137992142 cites W2622263826 @default.
- W3137992142 cites W2768489488 @default.
- W3137992142 cites W2795783309 @default.
- W3137992142 cites W2886335102 @default.
- W3137992142 cites W2888728082 @default.
- W3137992142 cites W2894651257 @default.
- W3137992142 cites W2935837427 @default.
- W3137992142 cites W2936599103 @default.
- W3137992142 cites W2949117887 @default.
- W3137992142 cites W2962721361 @default.
- W3137992142 cites W2962766617 @default.
- W3137992142 cites W2963037989 @default.
- W3137992142 cites W2963150697 @default.
- W3137992142 cites W2963351448 @default.
- W3137992142 cites W2963381188 @default.
- W3137992142 cites W2963403868 @default.
- W3137992142 cites W2963727650 @default.
- W3137992142 cites W2963815618 @default.
- W3137992142 cites W2963840672 @default.
- W3137992142 cites W2963849369 @default.
- W3137992142 cites W2963857746 @default.
- W3137992142 cites W2964241181 @default.
- W3137992142 cites W2964444661 @default.
- W3137992142 cites W2970575838 @default.
- W3137992142 cites W2982770724 @default.
- W3137992142 cites W2988452521 @default.
- W3137992142 cites W2989604896 @default.
- W3137992142 cites W3018757597 @default.
- W3137992142 cites W3034971973 @default.
- W3137992142 cites W3035396860 @default.
- W3137992142 cites W3039009902 @default.
- W3137992142 cites W3042011474 @default.
- W3137992142 cites W3106250896 @default.
- W3137992142 cites W3108849448 @default.
- W3137992142 cites W3116271762 @default.
- W3137992142 cites W3116756016 @default.
- W3137992142 cites W607748843 @default.
- W3137992142 doi "https://doi.org/10.48550/arxiv.2103.09460" @default.
- W3137992142 hasPublicationYear "2021" @default.
- W3137992142 type Work @default.
- W3137992142 sameAs 3137992142 @default.
- W3137992142 citedByCount "1" @default.
- W3137992142 crossrefType "posted-content" @default.
- W3137992142 hasAuthorship W3137992142A5000727470 @default.
- W3137992142 hasAuthorship W3137992142A5027961969 @default.
- W3137992142 hasAuthorship W3137992142A5061145810 @default.
- W3137992142 hasAuthorship W3137992142A5073212256 @default.
- W3137992142 hasAuthorship W3137992142A5080470862 @default.
- W3137992142 hasAuthorship W3137992142A5091093335 @default.
- W3137992142 hasBestOaLocation W31379921421 @default.
- W3137992142 hasConcept C105795698 @default.
- W3137992142 hasConcept C111919701 @default.
- W3137992142 hasConcept C118505674 @default.
- W3137992142 hasConcept C13280743 @default.
- W3137992142 hasConcept C138885662 @default.
- W3137992142 hasConcept C153180895 @default.
- W3137992142 hasConcept C154945302 @default.
- W3137992142 hasConcept C165064840 @default.
- W3137992142 hasConcept C177264268 @default.
- W3137992142 hasConcept C185798385 @default.
- W3137992142 hasConcept C199360897 @default.
- W3137992142 hasConcept C205649164 @default.
- W3137992142 hasConcept C2776151529 @default.
- W3137992142 hasConcept C2776401178 @default.
- W3137992142 hasConcept C2776760102 @default.
- W3137992142 hasConcept C2983787585 @default.
- W3137992142 hasConcept C33923547 @default.
- W3137992142 hasConcept C41008148 @default.
- W3137992142 hasConcept C41895202 @default.
- W3137992142 hasConcept C52622490 @default.
- W3137992142 hasConceptScore W3137992142C105795698 @default.
- W3137992142 hasConceptScore W3137992142C111919701 @default.
- W3137992142 hasConceptScore W3137992142C118505674 @default.
- W3137992142 hasConceptScore W3137992142C13280743 @default.
- W3137992142 hasConceptScore W3137992142C138885662 @default.