Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection

通过自适应训练样本选择弥补基于Anchor和无Anchor检测之间的差距.

Libra R-CNN: Towards Balanced Learning for Object Detection

Libra R-CNN: 面向目标检测中的均衡学习.

Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection

GFocal Loss: 为密集目标检测学习合格且分散的边界框.

图像长尾分布(Long-Tail Distribution)问题

Long-tail distribution problem in image datasets.

Every Model Learned by Gradient Descent Is Approximately a Kernel Machine

使用梯度下降优化的深度学习模型近似于核方法.

Unsupervised Adversarial Domain Adaptation for Micro-Doppler Based Human Activity Classification

将对抗域适应(ADA)应用于微多普勒人类活动分类.