Edwith 논문으로 짚어보는 딥러닝의 맥

모두의연구소 풀잎스쿨 6기 edwith 논문으로 보는 딥러닝의 맥을 보며 같이 공부하는 사람들이 정리한 repository 입니다.

이 자료들은 edwith 강의를 중심으로 해당 주제에 맞는 여러 참고 문헌들과 풀잎스쿨 발표자료를 정리한 것입니다.

• 전체 발표자들: 강성현, 강재호, 김경태, 김선호, 문동지, 이규희, 이일구, 조원양, 최성욱, 한상훈
• 퍼실이: 이일구

1주 (2019. 01. 12.)

• 4가지 CNN 살펴보기: AlexNet, VGG, GoogLeNet, ResNet
  • 발표자료
    • 강성현님: ResNet 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • ImageNet Classification with Deep Convolutional Neural Networks paper pdf
    • Very deep convolutional networks for large-scale image recognition https://arxiv.org/abs/1409.1556
    • Going deeper with convolutions https://arxiv.org/abs/1409.4842
    • Deep Residual Learning for Image Recognition https://arxiv.org/abs/1512.03385
  • 더 보면 좋을 참고문헌 (블로그 등)
    • Densely Connected Convolutional Networks https://arxiv.org/abs/1608.06993
    • 김성훈 교수님 DenseNet PR12
    • 유재준님 Inception and Xception PR12

2주 (2019. 01. 19.)

• Overfitting을 막는 regularization
  • 발표자료
    • 김경태님: 최성준님 발표자료 pptx link
    • 한상훈님: Batch Normalization 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • Dropout : A Simple Way to Prevent Neural Networks from Overfitting paper pdf
    • Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift https://arxiv.org/abs/1502.03167
  • 더 보면 좋을 참고문헌 (블로그 등)
    • 정영재님 Batch Normalization PR12

3주 (2019. 01. 26.)

• 이미지의 각 픽셀을 분류하는 Semantic Segmentation
  • 발표자료
    • 문동지님: Fully Convolutional Networks for Semantic Segmentation 발표자료 pdf link
    • 김선호님: DeepLab 발표자료 pdf link
    • 강재호님: UNet 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • Fully Convolutional Networks for Semantic Segmentation paper pdf
    • Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs https://arxiv.org/abs/1412.7062
    • Learning deconvolution network for semantic segmentation https://arxiv.org/abs/1505.04366
    • DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs https://arxiv.org/abs/1606.00915
  • 더 보면 좋을 참고문헌 (블로그 등)
    • 김태오님 DeepLab PR12

4주 (2019. 02. 09.)

• Residual Networks가 왜 잘 되는지 해석해보기
  • 발표자료
    • 이규희님: Resnet v2 (Identity Mappings in Deep Residual Networks) 발표자료 pdf link
    • 최성욱님: Exponential Ensembles 발표자료 pdf link
    • 조원양님: Wide ResNet 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • Deep Residual Learning for Image Recognition https://arxiv.org/abs/1512.03385
    • Identity Mappings in Deep Residual Networks https://arxiv.org/abs/1603.05027
    • Residual Networks are Exponential Ensembles of Relatively Shallow Networks https://arxiv.org/abs/1605.06431
    • Wide Residual Networks https://arxiv.org/abs/1605.07146

5주 (2019. 02. 16.)

• Image Detection 방법론: RCNN, SPPNet, FastRCNN, FasterRCNN
  • 발표자료
    • 이일구님: R-CNN, SPPNet (최성준님 팔표자료 이용) pptx link
    • 강성현님: Fast R-CNN, Faster R-CNN 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • Rich feature hierarchies for accurate object detection and semantic segmentation https://arxiv.org/abs/1311.2524
    • Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition https://arxiv.org/abs/1406.4729
    • Fast R-CNN https://arxiv.org/abs/1504.08083
    • Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks https://arxiv.org/abs/1506.01497
  • 더 보면 좋을 참고문헌 (블로그 등)
    • Oxford Visual Geometry Groups: Detection 자료 detection-part1.pdf, detection-part2.pdf
    • 이호성님 detection 정리 github
    • 김화평님 Faster R-CNN slideshare
    • 박진우님 Faster R-CNN blog
    • 이진원님 Faster R-CNN PR12

6주 (2019. 02. 23.)

• Image Detection 방법론: AttentionNet, SSD, YOLO YOLOv2
  • 발표자료
    • 문동지님: SSD, YOLO 발표자료 pdf link
    • 조원양님: RetinaNet 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • AttentionNet: Aggregating Weak Directions for Accurate Object Detection https://arxiv.org/abs/1506.07704
    • You Only Look Once: Unified, Real-Time Object Detection https://arxiv.org/abs/1506.02640
    • YOLO9000: Better, Faster, Stronger https://arxiv.org/abs/1612.08242
    • SSD: Single Shot MultiBox Detector https://arxiv.org/abs/1512.02325
  • 더 보면 좋을 참고문헌 (블로그 등)
    • 박진우님 YOLO blog
    • 이진원님 YOLO9000 PR12
    • YOLOv3: An Incremental Improvement https://arxiv.org/abs/1804.02767
    • 김태오님 Mask R-CNN PR12

7주 (2019. 03. 02.)

• 징검다리 휴일 (공부도 쉬는게 중요합니다)

8주 (2019. 03. 09.)

• 이미지와 질문이 주어졌을 때 답을 맞추는 Visual QnA
  • 발표자료
    • 이일구님: DPPNet (최성준님 발표자료 이용) pptx link
    • 한상훈님: Multimodal Compact Bilinear Pooling 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • Image Question Answering using Convolutional Neural Network with Dynamic Parameter Prediction https://arxiv.org/abs/1511.05756
    • Multimodal Compact Bilinear Pooling for Visual Question Answering and Visual Grounding https://arxiv.org/abs/1606.01847
  • 더 보면 좋을 참고문헌 (블로그 등)
    • DPPNet 논문에 나온 Wu-Palmer similarity youbute

9주 (2019. 03. 16.)

• 이미지를 설명하는 문장을 만들어내는 Image Captioning
  • 발표자료
    • 강재호님: Show and Tell / Show, Attend and Tell (최성준님 발표자료 이용) pptx link
    • 이규희님: DenseCap: Fully Convolutional Localization Networks for Dense Captioning 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • Show and Tell: Lessons learned from the 2015 MSCOCO Image Captioning Challenge
      • IEEE Transactions on Pattern Analysis and Machine Intelligence (Volume: 39 , Issue: 4 , April 1 2017)
      • https://arxiv.org/abs/1411.4555
    • Show, Attend and Tell: Neural Image Caption Generation with Visual Attention https://arxiv.org/abs/1502.03044
    • DenseCap: Fully Convolutional Localization Networks for Dense Captioning
      • Demo site
      • https://arxiv.org/abs/1511.07571
  • 더 보면 좋을 참고문헌 (github, 블로그 등)
    • Show and Tell TensorFlow official code TensorFlow models
    • Show, Attend and Tell TensorFlow official code TensorFlow Tutorials
    • DenseCap: Fully Convolutional Localization Networks for Dense Captioning author official code
    • 강지양님 Show and Tell PR12

10주 (2019. 03. 23.)

• 주어진 사진을 원하는 화풍으로 만드는 Neural Style
  • 발표자료
    • 문동지님: A Neural Algorithm of Artistic Style (최성준님 발표자료 이용) pptx link
    • 한상훈님: Perceptual losses for style transfer and super-resolution 발표자료 pdf link
    • 조원양님: Deep Photo Style Transfer 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • Texture Synthesis Using Convolutional Neural Networks
      • paper link
      • https://arxiv.org/abs/1505.07376
  • 더 보면 좋을 참고문헌 (github, 블로그 등)
    • Neural Style Transfer TensorFlow official code TensorFlow Tutorials
    • 김승일 소장님 Deep Photo Style Transfer PR12

11주 (2019. 03. 30.)

• Generative Adversarial Network
  • 발표자료
    • 김선호님: GAN, DCGAN, Pix2Pix, CycleGAN 발표자료 pdf link
  • 강의에서 제안한 참고문헌
    • Generative Adversarial Network https://arxiv.org/abs/1406.2661
    • Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks https://arxiv.org/abs/1511.06434
    • Generative Adversarial Text to Image Synthesis https://arxiv.org/abs/1605.05396
  • 더 보면 좋을 참고문헌 (github, 블로그 등)
    • NIPS 2016 Tutorial: Generative Adversarial Networks https://arxiv.org/abs/1701.00160
    • Generative Adversarial Networks : An Overview https://arxiv.org/abs/1710.07035
    • DCGAN TensorFlow official code TensorFlow Tutorials
    • Pix2Pix TensorFlow official code TensorFlow Tutorials
    • 이일구님 Generative models tensorflow version 2.0 style collection github
    • 이활석님 tensorflow-generative-model-collections github
    • 유재준님 GAN PR12
    • 김승일 소장님 GAN PR12
    • 차준범님 InfoGAN PR12