摘要： Read Data: 0909 Publication: Title: （多模态）Pixel-BERT: Aligning Image Pixels with Text by Deep Multi-Modal Transformers Aim: 从图像和句子对中，直接建立图像像素与语义之间更准确、更 阅读全文

摘要： Read Data:0908 Publication: ICML 2021 Title: Interpreting and Disentangling Feature Components of Various Complexity from DNNs Aim: 从一个新的角度分析深度神经网络的特征 阅读全文

摘要： Read Data: 0907 Publication: CVPR 2021 Title: You Only Look One-level Feature Aim: 保持简单、高效的原则，优化结构。 Research Question: 1、研究FPN的关键影响因素。 2、分而治之是很好的方法，但是 阅读全文

摘要： Read Data: 0906 Publication: CVPR 2021 Title: Less is More: CLIPBERT for Video-and-Language Learning via Sparse Sampling Aim: 解决视频-文本模型中，耗费大量空间和时间的问题。 阅读全文

摘要： Read Data: 0903 Publication: CVPR 2021 Title: Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition Aim: 本文提出的新型Patch- 阅读全文

摘要： Read Data: 0902 Publication: CVPR 2021 Title: Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detection 阅读全文

摘要： Read Data: 0901 Publication: CVPR 2021 Title: Dense Relation Distillation with Context-aware Aggregation for Few-Shot Object Detection Aim: 解决小样本对大量标注 阅读全文

摘要： Read Data: 0831 Publication: CVPR 2021 Title: Dynamic Head: Unifying Object Detection Heads with Attentions Aim: 本文试图在目标探测头呈现出统一的视图。 Research Question 阅读全文

摘要： Read Data: 0830 Publication: CVPR 2021 Title: RepVGG: Making VGG-style ConvNets Great Again Aim: 复杂的结构，虽然精度提高，但是伴随着一定问题：1、复杂的多分支设计使模型难以实现和定制，减慢了推理速度并降 阅读全文

摘要： Read Data: 0827 Publication: ICLR 2018 Title: Better to Follow, Follow to Be Better: Towards Precise Supervision of Feature Super-Resolution for Small 阅读全文