摘要:
1.Stand-Alone Self-Attention in Vision Models url:https://arxiv.org/abs/1906.05909 做了一个纯attention的网络,这里的attention是局部的,并且带了一个全局共享的relative position enc 阅读全文
摘要:
1.A Simple Framework for Contrastive Learning of Visual Representations url:https://arxiv.org/pdf/2002.05709.pdf 强Aug是做无监督的关键。Crop+Color jitter。搭配上Cot 阅读全文
摘要:
1.UNBIASED TEACHER FOR SEMI-SUPERVISED OBJECT DETECTION url:https://openreview.net/forum?id=MJIve1zgR_ code:https://github.com/facebookresearch/unbias 阅读全文
摘要:
单阶段目标检测 EfficientDet:EfficientDet: Scalable and Efficient Object Detection url:https://arxiv.org/abs/1911.09070 引入了BiFPN,作者发现PANet涨点挺猛的。然后就在上面改一下,而且不能 阅读全文
摘要:
分类: 1.Test-Agnostic Long-Tailed Recognition by Test-Time Aggregating Diverse Experts with Self-Supervision url:https://arxiv.org/pdf/2107.09249.pdf 大家 阅读全文
摘要:
待补充 阅读全文