End to End Object Detection with Transformer

Transcript

1. ͲΜͳ΋ͷʁ ઌߦݚڀͱൺ΂ͯԿ͕͍͢͝ʁ ٕज़ͷख๏΍؊͸ʁ վળͷ༨஍͸͋Δʁ Ͳ͏΍ͬͯ༗ޮͩͱݕূͨ͠ʁ ࣍ʹಡΉ΂͖࿦จ͸ʁ ɾLin, T.Y., Dolla ́r, P., Girshick, R., He, K., Hariharan, B., Belongie, S.: Feature pyramid networks for object detection. In: CVPR (2017) ɾVaswani,A.,Shazeer,N.,Parmar,N.,Uszkoreit,J.,Jones,L.,Gomez,A.N.,Kaiser, L., Polosukhin, I.: Attention is all you need. In: NeurIPS (2017) ɾը૾தͷେ͖͍෺ମݕग़ʹ͸ޮՌత͕ͩখ͍͞෺ମݕग़ʹ͍ͭͯ͸௿͍ύϑΥʔϚϯ εͱͳͬͯ͠·͍ͬͯΔɻ ɾCOCO 2017 detection and panoptic segmentation datasets Λ༻͍࣮ͯݧ ɾ࠷దԽ͞ΕͨcFaster - RCNNͱಉ౳ͷύϑΥʔϚϯεͰ͋ΔࣄΛݕূ ɾಛʹେ͖ͳ෺ମݕ஌ʹ͓͍ͯ͸طଘख๏Ҏ্ͷύϑΥʔϚϯεΛൃش ɾॲཧ࣌ؒ΋໿൒෼ͰࡁΉ༷ʹ ɾ༩͑ΒΕͨը૾ʹ͍ࣸͬͯΔݻ༗ͳ෺ମͷਖ਼ղͷ෺ମΫϥεɺҐஔɺαΠζͱɺ༧ଌ ͞Εͨ͜ΕΒͷ஋͔Βܭࢉ͞ΕͨଛࣦΛ࠷దԽ͢Δɻ ɾ෺ମத৺ͷάϦουʹରͯ͠༧ଌΛߦ͏Ұஈ֊༧ଌ΍෺ମྖҬͷఏҊʹରͯ͠ϘοΫ εΛ༧ଌ͢Δೋஈ֊ݕग़ͷϓϩηεΛഉআ͠ɺೖྗը૾ʹର͢Δઈରശ༧ଌͰݕग़ηο τΛ௚઀༧ଌ͢Δ͜ͱͰɺݕग़ϓϩηεΛޮ཰Խ ɾDEtection TRansformerʢDETR)͸Transformerͱbipartile matching lossΛ༻͍ͯ෺ମΛݕ஌͢Δ৽͍͠ϑϨʔϜϫʔΫ End-to-End Object Detection with Transformers https://arxiv.org/pdf/2005.12872.pdf ʢ2020ʣNicolas Carion⋆, Francisco Massa⋆, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, and Sergey Zagoruyko Facebook AI , 2021/03/24
2. ࣍ʹಡΉ࿦จଓ͖ ɾHe, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016)