导入相关库 In [15]: from __future__ import division
from numpy.random import randn
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
plt.rc(figure, figsize(12, 5))
np.set_printoptions(precision4)
%pwdOut[15]: uD:\\ipython 加载数据 In [4…
Fine-grained Detection —— TransFG(2022.02.23) 1. Part Selection Module2. Contrastive Feature Learning3. My Thinking3.1. PSM部分3.2. CFL部分 4. My Summary 文章:TransFG: A Transformer Architecture for Fine-grained Recogniti…
【补充精读】Supplement for “ Fine-Grained Face Swapping via Regional GAN Inversion” 0、前言1. Shape Swapping Details2. Loss Functions2.1 Pixel-wise reconstruction loss.2.2 Multi-scale LPIPS loss.2.3 Multi-scale face inversion loss.2.4 Adversarial loss.3.…
Fine-grained Detection —— DCL(2022.02.18) 1. Region Confusion Mechanism2. Region Alignment Network3. My Thinking3.1. RCM部分3.2. RAN部分 4. My Summary 文章:Destruction and Construction Learning for Fine-grained Image Reco…
文章目录 基础信息Abstract1 Introduction2 Related Work3 Noisy Fine-Grained Data3.1 Categories3.2 Images from the Web : Web 数据的获取方式 4 Data via Active Learning5 Experiments6 Discussion 基础信息
The Unreasonable Effectiveness of Noisy Data f…
EMNLP2021 目录 IntroductionProblem FormulationC2F FrameworkInitial Fine-grained Weak SupervisionTailored Language Model TrainingHierarchy-Aware RegularizationPseudo Training Data Generation, Text Classifier, & Weak Supervision UpdateExperiments参考文献…
AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networks 一、摘要二、AttnGAN优势三、两大核心四、架构分析注意力生成网络注意力模型 F a t t n ( e , h ) F^{attn}(e,h) Fattn(e,h)损失函数 DAMSM文本编码器图像编码器注意力驱动…