Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312305279> ?p ?o ?g. }
- W4312305279 endingPage "18" @default.
- W4312305279 startingPage "1" @default.
- W4312305279 abstract "Current object detection systems and benchmarks typically handle a limited number of categories, up to about a thousand categories. This paper scales the number of categories for object detection systems and benchmarks up to 21,000, by leveraging existing object detection and image classification data. Unlike previous efforts that usually transfer knowledge from base detectors to image classification data, we propose to rely more on a reverse information flow from a base image classifier to object detection data. In this framework, the large-vocabulary classification capability is first learnt thoroughly using only the image classification data. In this step, the image classification problem is reformulated as a special configuration of object detection that treats the entire image as a special RoI. Then, a simple multi-task learning approach is used to join the image classification and object detection data, with the backbone and the RoI classification branch shared between two tasks. This two-stage approach, though very simple without a sophisticated process such as multi-instance learning (MIL) to generate pseudo labels for object proposals on the image classification data, performs rather strongly that it surpasses previous large-vocabulary object detection systems on a standard evaluation protocol of tailored LVIS. Considering that the tailored LVIS evaluation only accounts for a few hundred novel object categories, we present a new evaluation benchmark that assesses the detection of all 21,841 object classes in the ImageNet-21K dataset. The baseline approach and evaluation benchmark will be publicly available at https://github.com/SwinTransformer/Simple-21K-Detection. We hope these would ease future research on large-vocabulary object detection." @default.
- W4312305279 created "2023-01-04" @default.
- W4312305279 creator A5014838804 @default.
- W4312305279 creator A5015898666 @default.
- W4312305279 creator A5025592561 @default.
- W4312305279 creator A5028650738 @default.
- W4312305279 creator A5032447166 @default.
- W4312305279 creator A5048295582 @default.
- W4312305279 creator A5073435344 @default.
- W4312305279 creator A5088888083 @default.
- W4312305279 date "2022-01-01" @default.
- W4312305279 modified "2023-10-16" @default.
- W4312305279 title "A Simple Approach and Benchmark for 21,000-Category Object Detection" @default.
- W4312305279 cites W1861492603 @default.
- W4312305279 cites W2031489346 @default.
- W4312305279 cites W2061629163 @default.
- W4312305279 cites W2088049833 @default.
- W4312305279 cites W2102605133 @default.
- W4312305279 cites W2108598243 @default.
- W4312305279 cites W2194775991 @default.
- W4312305279 cites W2277195237 @default.
- W4312305279 cites W2295107390 @default.
- W4312305279 cites W2331143823 @default.
- W4312305279 cites W2560647685 @default.
- W4312305279 cites W2565639579 @default.
- W4312305279 cites W2570343428 @default.
- W4312305279 cites W2604260814 @default.
- W4312305279 cites W2813911573 @default.
- W4312305279 cites W2883554151 @default.
- W4312305279 cites W2940046420 @default.
- W4312305279 cites W2948672349 @default.
- W4312305279 cites W2950557962 @default.
- W4312305279 cites W2962685835 @default.
- W4312305279 cites W2963045696 @default.
- W4312305279 cites W2963150697 @default.
- W4312305279 cites W2963603913 @default.
- W4312305279 cites W2963936013 @default.
- W4312305279 cites W2964274719 @default.
- W4312305279 cites W2982770724 @default.
- W4312305279 cites W2983943451 @default.
- W4312305279 cites W2992308087 @default.
- W4312305279 cites W2997044845 @default.
- W4312305279 cites W2998508940 @default.
- W4312305279 cites W3035682985 @default.
- W4312305279 cites W3109576564 @default.
- W4312305279 cites W3110272085 @default.
- W4312305279 cites W3138516171 @default.
- W4312305279 cites W3142837074 @default.
- W4312305279 cites W3173859428 @default.
- W4312305279 cites W3176659256 @default.
- W4312305279 cites W3176774696 @default.
- W4312305279 cites W3190805089 @default.
- W4312305279 cites W4288083516 @default.
- W4312305279 cites W4312424618 @default.
- W4312305279 cites W7746136 @default.
- W4312305279 doi "https://doi.org/10.1007/978-3-031-20083-0_1" @default.
- W4312305279 hasPublicationYear "2022" @default.
- W4312305279 type Work @default.
- W4312305279 citedByCount "0" @default.
- W4312305279 crossrefType "book-chapter" @default.
- W4312305279 hasAuthorship W4312305279A5014838804 @default.
- W4312305279 hasAuthorship W4312305279A5015898666 @default.
- W4312305279 hasAuthorship W4312305279A5025592561 @default.
- W4312305279 hasAuthorship W4312305279A5028650738 @default.
- W4312305279 hasAuthorship W4312305279A5032447166 @default.
- W4312305279 hasAuthorship W4312305279A5048295582 @default.
- W4312305279 hasAuthorship W4312305279A5073435344 @default.
- W4312305279 hasAuthorship W4312305279A5088888083 @default.
- W4312305279 hasConcept C115961682 @default.
- W4312305279 hasConcept C119857082 @default.
- W4312305279 hasConcept C124101348 @default.
- W4312305279 hasConcept C13280743 @default.
- W4312305279 hasConcept C138885662 @default.
- W4312305279 hasConcept C153180895 @default.
- W4312305279 hasConcept C154945302 @default.
- W4312305279 hasConcept C182521987 @default.
- W4312305279 hasConcept C185798385 @default.
- W4312305279 hasConcept C205649164 @default.
- W4312305279 hasConcept C2776151529 @default.
- W4312305279 hasConcept C2777601683 @default.
- W4312305279 hasConcept C2781238097 @default.
- W4312305279 hasConcept C31510193 @default.
- W4312305279 hasConcept C31972630 @default.
- W4312305279 hasConcept C41008148 @default.
- W4312305279 hasConcept C41895202 @default.
- W4312305279 hasConcept C4641261 @default.
- W4312305279 hasConcept C64876066 @default.
- W4312305279 hasConcept C71681937 @default.
- W4312305279 hasConcept C75294576 @default.
- W4312305279 hasConcept C95623464 @default.
- W4312305279 hasConceptScore W4312305279C115961682 @default.
- W4312305279 hasConceptScore W4312305279C119857082 @default.
- W4312305279 hasConceptScore W4312305279C124101348 @default.
- W4312305279 hasConceptScore W4312305279C13280743 @default.
- W4312305279 hasConceptScore W4312305279C138885662 @default.
- W4312305279 hasConceptScore W4312305279C153180895 @default.
- W4312305279 hasConceptScore W4312305279C154945302 @default.
- W4312305279 hasConceptScore W4312305279C182521987 @default.