PTCA635
发表于 2025-3-30 10:32:45
Spyros Blavoukos,Dimitris Bourantonists in large-scale detection benchmarks (. the COCO dataset), the performance on small objects is far from satisfactory. The reason is that small objects lack sufficient detailed appearance information, which can distinguish them from the background or similar objects. To deal with the small object d
EPT
发表于 2025-3-30 12:46:17
http://reply.papertrans.cn/24/2342/234196/234196_52.png
Pelago
发表于 2025-3-30 18:59:51
http://reply.papertrans.cn/24/2342/234196/234196_53.png
arrogant
发表于 2025-3-30 21:45:33
Alexander Antonov,Tanel Kerikmäepublicly available large diverse datasets. In this paper, we introduce CCPD, a large and comprehensive LP dataset. All images are taken manually by workers of a roadside parking management company and are annotated carefully. To our best knowledge, CCPD is the largest publicly available LP dataset t
一大块
发表于 2025-3-31 00:57:16
http://reply.papertrans.cn/24/2342/234196/234196_55.png
lanugo
发表于 2025-3-31 07:09:23
Lecture Notes in Computer Sciencehttp://image.papertrans.cn/c/image/234196.jpg
改良
发表于 2025-3-31 09:22:06
http://reply.papertrans.cn/24/2342/234196/234196_57.png
女歌星
发表于 2025-3-31 14:43:07
http://reply.papertrans.cn/24/2342/234196/234196_58.png
大洪水
发表于 2025-3-31 18:53:40
https://doi.org/10.1007/978-1-84800-171-8 using a batch size of 2; when using typical batch sizes, GN is comparably good with BN and outperforms other normalization variants. Moreover, GN can be naturally transferred from pre-training to fine-tuning. GN can outperform its BN-based counterparts for object detection and segmentation in COCO,
Mangle
发表于 2025-4-1 01:26:33
http://reply.papertrans.cn/24/2342/234196/234196_60.png