Scaled-YOLOv4: Scaling Cross Stage Partial Network In this reading notes: We have reviewed some basic model scaling method: width, depth, resolution, compound scaling. We have computed the operation amount of residual blocks, and showed the relation with input image size (square), number of layers (linear), number of filters (square). We…
Paper Information Paper: YOLObile: Real-Time Object Detection on Mobile Devices via Compression-Compilation Co-Design Authors: Yuxuan Cai, Hongjia Li, Geng Yuan, Wei Niu, Yanyu Li, Xulong Tang, Bin Ren, Yanzhi Wang Paper: https://arxiv.org/abs/2009.05697 Github: https://github.com/nightsnack/YOLObile Objective: Real-time object detection for mobile devices. Study notes and presentation: Download: https://connectpolyu-my.sharepoint.com/:p:/g/personal/18048204r_connect_polyu_hk/EcRbix5iqshBglmxuLurS-sBBFmbrk8chRkim1y54-yOXw?e=8Qdfmd This is an…
Abstract The paper introduces a position attention module and a channel attention module to capture global dependencies in the spatial and channel dimensions respectively. The proposed DANet adaptively integrates local semantic features using the self-attention mechanism. 摘要 本文引入了位置关注模块和通道关注模块，分别在空间和通道维度上捕捉全局依赖性。 所提出的DANet利用自注意力机制自适应地集成局部语义特征。 Outline Brief Review: attention mechanism, SE net DANet: Dual Attention Network…
Objectives Deep learning is a recently hot machine learning method. The deep learning architectures are formed by the composition of several nonlinear transformations with the goal to yield more abstract and extract useful representations/features. (i) Start with a revision of the basic principle of Neural Networks, neutron structure, examples of…
Objectives Deep learning is a recently hot machine learning method. The deep learning architectures are formed by the composition of several nonlinear transformations with the goal to yield more abstract and extract useful representations/features. (i) Start with a revision of the basic principle of Neural Networks, neutron structure, examples of…