'2019/10'에 해당되는 글 78건

  1. 2019.10.31 From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter f..
  2. 2019.10.31 안녕하세요! 어느덧 내일이 ICCV main conference 마지막 날이네요. 내일 오전 10시 30분에 143번에서 tag2pix poster 발표를 합니다. Color tag를 이용해서 스케치를 자동으로 채색하는 논문인데, 관심 있으..
  3. 2019.10.31 https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw/videos 여기에 CVPR 영상들이 있는데, 19년도 튜토리얼은 안 보이네요.. https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw/playlists 18년도 튜토리얼은 수록되어 ..
  4. 2019.10.31 머신러닝 모델 디버깅 리소스와 팁 https://t.co/9Y7kDc1hag?amp=1 https://medium.com/infinity-aka-aseem/things-we-wish-we-had-known-before-we-started-our-first-machine-learning-project-336d1d6f2184 https://medium.com/@keeper6928/how-to-uni..
  5. 2019.10.31 혹시 ICCV 2019 영상 올라오나요? https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw 여기에 튜토리얼이랑 메인컨퍼런스 구두발표는 아마 올라올거에요~
  6. 2019.10.31 From Google brain researchers @NeurIPS 2019: Learning to Predict Without Looking Ahead https://www.profillic.com/paper/arxiv:1910.13038 "Rather than hardcoding forward prediction, we try to get agents to *learn* that they need to predict the future"
  7. 2019.10.31 Component Attention Guided Face Super-Resolution Network: CAGFace 얼굴에 특화된 4배 확대 SR 신경망 모델인데... 성능이 상당히 좋네요. 대신 신경망도 덩치가 크네요. 학습 파라메터가 6천만개가 넘습니다. ..
  8. 2019.10.31 This video gives a quick overview of 41 research papers presented by Google at the International Conference on Computer Vision (ICCV)! [https://youtu.be/z-yvY8iAaHM](https://t.co/1q6od2KUzp?amp=1)
  9. 2019.10.30 안녕하세요, TensorFlow KR 여러분, 저는 현재 AI를 활용한 이미지 변형을 이용해 작업을 제작 중에 있는 미술학도입니다. TensorFlow KR 여러분들에게 제가 진행중인 작업에 관해 조언을 받고 싶어 ..
  10. 2019.10.30 #Download #Source #Code from Video Description. This is complete zero to Amazon Review Sentiment Classification Lesson. In this lesson, I will discuss the following sections 1. What is NLP 2. Applications of NLP 3. Text Data Cleaning Options 4. Bag of W..
  11. 2019.10.30 Learn: 1. linear algebra well (e.g. matrix math) 2. calculus to an ok level (not advanced stuff) 3. prob. theory and stats to a good level 4. theoretical computer science basics 5. to code well in Python and ok in C++ Then read and implement ML papers..
  12. 2019.10.29 Some of the best courses available on the internet. 1. Data Science Professional Certificate from IBM. http://bit.ly/2M6o4ko 2. Best Machine Learning course anyone can find on the Internet via Andrew NG/Stanford. http://bit.ly/2J9uZWV 3. Persona..
  13. 2019.10.29 Deep Learning Drizzle Drench yourself in Deep Learning, Reinforcement Learning, Machine Learning, Computer Vision, and NLP by learning from these exciting lectures!! GitHub by Marimuthu Kalimuthu: https://github.com/kmario23/deep-learning-drizzle We..
  14. 2019.10.29 From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter ..
  15. 2019.10.29 #정보공유 #행사 안녕하세요! RLKorea 운영진입니다! 지난 10월 27~28일 RLKorea Bootcamp가 진행되었는데요! 강화학습의 기초개념인 MDP부터 시작하여 DQN, A2C, DDPG, SAC 등 다양한 강화학습 알고리즘..
  16. 2019.10.29 "Prior가 뭔지는 알잖아요?" 동의하는척 웃고 넘어간 뒤 한 없이 부끄러워져서 머신러닝 관련 기본적인 통계추론 내용을 정리를 해보았습니다. 논문 읽을 때 자주 나오는 내용인데도 생각보다 ..
  17. 2019.10.29 Check out these Top #MachineLearning Youtube Videos Under 10 Minutes
  18. 2019.10.29 Turn a line sketch into a photorealistic face: https://www.profillic.com/paper/arxiv:1910.08914 From sparse lines that coarsely describe a face, photorealistic images can be generated using conditional self-attention generative adversarial network (CS..
  19. 2019.10.28 [XGBoost/LightGBM] Laurae++: xgboost / LightGBM 이번에는 Kaggle에서 가장 많이 사용되는 모델인 xgboost와 lgbm에 관련된 내용입니다. @laurae 님이 만든 xgboost/lightgbm 웹페이지입니다. 공식 documentation에서도 ..
  20. 2019.10.28 DEVIEW2019 Keynote에서 “석상옥 대표님”이 소개해주신 NAVER LABS의 자율주행용 Open dataset입니다. 국내자율주행 기술 성장에 큰 도움이 될 것으로 기대됩니다 : )
  21. 2019.10.28 안녕하세요 TmaxData에서 NLP를 연구 중인 장영록입니다:) ALBERT(A Lite BERT - Google 2019.9)라는 논문을 소개드리고자 글 적 습니다. ALBERT는 BERT 보다 모델의 크기는 작지만 GLUE, SQuAD 등의 task에서 더 ..
  22. 2019.10.28 안녕하세요 rl kr. 개인적으로 오랜 숙제였던 IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures(IMPALA)를 구현하여 결과를 공유합니다. tensorflow로 구현하였습니다(pytorch로 하지..
  23. 2019.10.27 In this lesson, we will discuss how you can expand the NLP model in SpaCy. Text data is so wide therefore it is possible that the model is not present in Spacy to deal with it and in that case you can build your own rule and merge with the default SpaCy..
  24. 2019.10.27 #Download working file from the video description Processing Pipeline is an important step in feature extraction from text data. Learn how the pipeline works and how you can speed up the processing by disabling some blocks of the pipeline. NLP Tutorial ..
  25. 2019.10.27 State of the art in Brain Tumor Segmentation: https://www.profillic.com/paper/arxiv:1909.12901 (A 3D U-net based deep learning model has been trained with the help of brain-wise normalization and patching strategies for the brain tumor segmentation ta..
  26. 2019.10.27 정말 대단함! 일독을 권합니다~
  27. 2019.10.27 Deep Learning with Keras A-Z: Part of our Data Science & Machine Learning Notes Series. We will now explore step by step tutor using Keras on MNIST. Go through the Analysis steps and try to understand the context. If some topics are too advanced they wi..
  28. 2019.10.26 ICYMI from BMVC 2019: human motion transfer - generation of a video https://www.profillic.com/paper/arxiv:1910.09139 (Their GAN-based architecture, DwNet, leverages dense intermediate pose-guided representation and refinement process to warp the requi..
  29. 2019.10.26 From ICCV: Gaze360, a large-scale gaze-tracking dataset and method for robust 3D gaze estimation in unconstrained images https://www.profillic.com/paper/arxiv:1910.10088
  30. 2019.10.25 State of the art in object detection: FuseMODNet: Real-Time Camera and LiDAR based Moving Object Detection for robust low-light Autonomous Driving https://www.profillic.com/paper/arxiv:1910.05395
From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video
https://www.profillic.com/paper/arxiv:1910.06407

FireNet: Real-time Segmentation of Fire Perimeter from Aerial Video
https://www.facebook.com/groups/DeepNetGroup/permalink/987081835018032/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

PR12 논문읽기 모임의 204번째 논문발표 입니다. https://www.youtube.com/watch?v=YNicvevmByo&feature=youtu.be ICLR2019 에서 발표된 "Learning deep representations by mutual information estimation and maximization" 이라는 논문을 ..  (0) 2019.11.04
안녕하세요, 전 수아랩 현 코그넥스 에서 머신러닝 엔지니어로 일하고있는 이호성이라고합니다. 지난주 서울에서 열린 ICCV 2019 학회에 다녀온 후기와, Best Paper로 선정된 “SinGAN: Learning a Genera..  (0) 2019.11.04
From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter f..  (0) 2019.10.31
안녕하세요! 어느덧 내일이 ICCV main conference 마지막 날이네요. 내일 오전 10시 30분에 143번에서 tag2pix poster 발표를 합니다. Color tag를 이용해서 스케치를 자동으로 채색하는 논문인데, 관심 있으..  (0) 2019.10.31
From Google brain researchers @NeurIPS 2019: Learning to Predict Without Looking Ahead https://www.profillic.com/paper/arxiv:1910.13038 "Rather than hardcoding forward prediction, we try to get agents to *learn* that they need to predict the future"  (0) 2019.10.31
This video gives a quick overview of 41 research papers presented by Google at the International Conference on Computer Vision (ICCV)! [https://youtu.be/z-yvY8iAaHM](https://t.co/1q6od2KUzp?amp=1)  (0) 2019.10.31
Posted by saveone

댓글을 달아 주세요

안녕하세요! 어느덧 내일이 ICCV main conference 마지막 날이네요. 내일 오전 10시 30분에 143번에서 tag2pix poster 발표를 합니다.

Color tag를 이용해서 스케치를 자동으로 채색하는 논문인데, 관심 있으신 분들은 오셔서 같이 이야기 나누었으면 좋겠습니다. 저는 GAN, detection, domain adaptation 등에 관심이 많습니다 ㅎㅎ

코드와 데이터셋 배포했습니다. 감사합니다 :)

Paper: https://arxiv.org/abs/1908.05840
Code: https://github.com/blandocs/Tag2Pix
GUI: https://github.com/MerHS/tag2pix-gui
https://www.facebook.com/groups/TensorFlowKR/permalink/1024410971233294/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

안녕하세요, 전 수아랩 현 코그넥스 에서 머신러닝 엔지니어로 일하고있는 이호성이라고합니다. 지난주 서울에서 열린 ICCV 2019 학회에 다녀온 후기와, Best Paper로 선정된 “SinGAN: Learning a Genera..  (0) 2019.11.04
From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter f..  (0) 2019.10.31
안녕하세요! 어느덧 내일이 ICCV main conference 마지막 날이네요. 내일 오전 10시 30분에 143번에서 tag2pix poster 발표를 합니다. Color tag를 이용해서 스케치를 자동으로 채색하는 논문인데, 관심 있으..  (0) 2019.10.31
From Google brain researchers @NeurIPS 2019: Learning to Predict Without Looking Ahead https://www.profillic.com/paper/arxiv:1910.13038 "Rather than hardcoding forward prediction, we try to get agents to *learn* that they need to predict the future"  (0) 2019.10.31
This video gives a quick overview of 41 research papers presented by Google at the International Conference on Computer Vision (ICCV)! [https://youtu.be/z-yvY8iAaHM](https://t.co/1q6od2KUzp?amp=1)  (0) 2019.10.31
From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter ..  (0) 2019.10.29
Posted by saveone

댓글을 달아 주세요

CVPR 2019 Tutorial, 컴퓨터 비전을 위한 심층강화학습
http://ivg.au.tsinghua.edu.cn/DRLCV/

튜토리얼 슬라이드
http://ivg.au.tsinghua.edu.cn/DRLCV/CVPR19_tutorial.pdf




https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw/videos 여기에 CVPR 영상들이 있는데, 19년도 튜토리얼은 안 보이네요.. https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw/playlists 18년도 튜토리얼은 수록되어 있는 것으로 보아 나중에 올라올지 모르겠습니다.

'Deep Learning > course' 카테고리의 다른 글

이번 ICCV 자료를 보다가, 예전 ICCV / CVPR 영상들을 보니까 좋은 워크샵 / 튜토리얼 영상들이 있더라구요! 컴퓨터 비전 공부하시는 분들께 도움되는 자료가 많은 것 같아 공유합니다 😇 ICCV 2019 ..  (0) 2019.11.27
A complete list of six video lectures in Generative adversarial network (GAN) is available on my YouTube Channel https://www.youtube.com/playlist?list=PLdxQ7SoCLQAMGgQAIAcyRevM8VvygTpCu You can subscribe my channel for more such videos https://www.yo..  (0) 2019.11.05
https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw/videos 여기에 CVPR 영상들이 있는데, 19년도 튜토리얼은 안 보이네요.. https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw/playlists 18년도 튜토리얼은 수록되어 ..  (0) 2019.10.31
혹시 ICCV 2019 영상 올라오나요? https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw 여기에 튜토리얼이랑 메인컨퍼런스 구두발표는 아마 올라올거에요~  (0) 2019.10.31
Learn: 1. linear algebra well (e.g. matrix math) 2. calculus to an ok level (not advanced stuff) 3. prob. theory and stats to a good level 4. theoretical computer science basics 5. to code well in Python and ok in C++ Then read and implement ML papers..  (0) 2019.10.30
Some of the best courses available on the internet. 1. Data Science Professional Certificate from IBM. http://bit.ly/2M6o4ko 2. Best Machine Learning course anyone can find on the Internet via Andrew NG/Stanford. http://bit.ly/2J9uZWV 3. Persona..  (0) 2019.10.29
Posted by saveone

댓글을 달아 주세요

머신러닝 모델 디버깅 리소스와 팁

https://t.co/9Y7kDc1hag?amp=1

https://medium.com/infinity-aka-aseem/things-we-wish-we-had-known-before-we-started-our-first-machine-learning-project-336d1d6f2184

https://medium.com/@keeper6928/how-to-unit-test-machine-learning-code-57cf6fd81765

https://pcc.cs.byu.edu/2017/10/02/practical-advice-for-building-deep-neural-networks/amp/?__twitter_impression=true

https://medium.com/ai%C2%B3-theory-practice-business/top-6-errors-novice-machine-learning-engineers-make-e82273d394db

http://karpathy.github.io/2019/04/25/recipe/

https://github.com/EricSchles/drifter_ml
https://www.facebook.com/303538826748786/posts/804661033303227/?sfnsn=mo

'Deep Learning > resources' 카테고리의 다른 글

[Model Interpretation] Deep Dive into Catboost Functionalities for Model Interpretation --- Tree기반 모델은 Feature Importance를 수치화해서 보여주지만 여전히 해석에 있어서는 무리가 있습니다. 수치도 기준에 따라 ..  (0) 2019.11.05
[서버의 VS Code를 띄워놓고 랩탑, 태블릿, 스마트폰에서 코딩하는 방법] 안녕하세요! 지난 겨울 docker로 원격작업하는 방법을 공유했던 연세대학교 통합과정 이응빈입니다. 어제 iPad를 새로 ..  (0) 2019.11.05
머신러닝 모델 디버깅 리소스와 팁 https://t.co/9Y7kDc1hag?amp=1 https://medium.com/infinity-aka-aseem/things-we-wish-we-had-known-before-we-started-our-first-machine-learning-project-336d1d6f2184 https://medium.com/@keeper6928/how-to-uni..  (0) 2019.10.31
#Download #Source #Code from Video Description. This is complete zero to Amazon Review Sentiment Classification Lesson. In this lesson, I will discuss the following sections 1. What is NLP 2. Applications of NLP 3. Text Data Cleaning Options 4. Bag of W..  (0) 2019.10.30
#정보공유 #행사 안녕하세요! RLKorea 운영진입니다! 지난 10월 27~28일 RLKorea Bootcamp가 진행되었는데요! 강화학습의 기초개념인 MDP부터 시작하여 DQN, A2C, DDPG, SAC 등 다양한 강화학습 알고리즘..  (0) 2019.10.29
[XGBoost/LightGBM] Laurae++: xgboost / LightGBM 이번에는 Kaggle에서 가장 많이 사용되는 모델인 xgboost와 lgbm에 관련된 내용입니다. @laurae 님이 만든 xgboost/lightgbm 웹페이지입니다. 공식 documentation에서도 ..  (0) 2019.10.28
Posted by saveone

댓글을 달아 주세요

혹시 ICCV 2019 영상 올라오나요? Jitendra Malik이 한 토크가 궁금해서 보고싶은데 못 찾겠네요 ㅠㅠ

https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw

여기에 튜토리얼이랑 메인컨퍼런스 구두발표는 아마 올라올거에요~
https://www.facebook.com/groups/TensorFlowKR/permalink/1023962494611475/?sfnsn=mo

'Deep Learning > course' 카테고리의 다른 글

A complete list of six video lectures in Generative adversarial network (GAN) is available on my YouTube Channel https://www.youtube.com/playlist?list=PLdxQ7SoCLQAMGgQAIAcyRevM8VvygTpCu You can subscribe my channel for more such videos https://www.yo..  (0) 2019.11.05
https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw/videos 여기에 CVPR 영상들이 있는데, 19년도 튜토리얼은 안 보이네요.. https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw/playlists 18년도 튜토리얼은 수록되어 ..  (0) 2019.10.31
혹시 ICCV 2019 영상 올라오나요? https://www.youtube.com/channel/UC0n76gicaarsN_Y9YShWwhw 여기에 튜토리얼이랑 메인컨퍼런스 구두발표는 아마 올라올거에요~  (0) 2019.10.31
Learn: 1. linear algebra well (e.g. matrix math) 2. calculus to an ok level (not advanced stuff) 3. prob. theory and stats to a good level 4. theoretical computer science basics 5. to code well in Python and ok in C++ Then read and implement ML papers..  (0) 2019.10.30
Some of the best courses available on the internet. 1. Data Science Professional Certificate from IBM. http://bit.ly/2M6o4ko 2. Best Machine Learning course anyone can find on the Internet via Andrew NG/Stanford. http://bit.ly/2J9uZWV 3. Persona..  (0) 2019.10.29
Deep Learning Drizzle Drench yourself in Deep Learning, Reinforcement Learning, Machine Learning, Computer Vision, and NLP by learning from these exciting lectures!! GitHub by Marimuthu Kalimuthu: https://github.com/kmario23/deep-learning-drizzle We..  (0) 2019.10.29
Posted by saveone

댓글을 달아 주세요

From Google brain researchers @NeurIPS 2019: Learning to Predict Without Looking Ahead

https://www.profillic.com/paper/arxiv:1910.13038

"Rather than hardcoding forward prediction, we try to get agents to *learn* that they need to predict the future"
https://www.facebook.com/groups/1738168866424224/permalink/2439511069623330/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter f..  (0) 2019.10.31
안녕하세요! 어느덧 내일이 ICCV main conference 마지막 날이네요. 내일 오전 10시 30분에 143번에서 tag2pix poster 발표를 합니다. Color tag를 이용해서 스케치를 자동으로 채색하는 논문인데, 관심 있으..  (0) 2019.10.31
From Google brain researchers @NeurIPS 2019: Learning to Predict Without Looking Ahead https://www.profillic.com/paper/arxiv:1910.13038 "Rather than hardcoding forward prediction, we try to get agents to *learn* that they need to predict the future"  (0) 2019.10.31
This video gives a quick overview of 41 research papers presented by Google at the International Conference on Computer Vision (ICCV)! [https://youtu.be/z-yvY8iAaHM](https://t.co/1q6od2KUzp?amp=1)  (0) 2019.10.31
From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter ..  (0) 2019.10.29
Turn a line sketch into a photorealistic face: https://www.profillic.com/paper/arxiv:1910.08914 From sparse lines that coarsely describe a face, photorealistic images can be generated using conditional self-attention generative adversarial network (CS..  (0) 2019.10.29
Posted by saveone

댓글을 달아 주세요

Component Attention Guided Face Super-Resolution Network: CAGFace

얼굴에 특화된 4배 확대 SR 신경망 모델인데...
성능이 상당히 좋네요.

대신 신경망도 덩치가 크네요. 학습 파라메터가 6천만개가 넘습니다.

https://arxiv.org/pdf/1910.08761.pdf
https://www.facebook.com/groups/TensorFlowKR/permalink/1024154381258953/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

This video gives a quick overview of 41 research papers presented by Google at the International Conference on Computer Vision (ICCV)!

 [https://youtu.be/z-yvY8iAaHM](https://t.co/1q6od2KUzp?amp=1)
https://www.facebook.com/groups/DeepNetGroup/permalink/989370031455879/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

안녕하세요! 어느덧 내일이 ICCV main conference 마지막 날이네요. 내일 오전 10시 30분에 143번에서 tag2pix poster 발표를 합니다. Color tag를 이용해서 스케치를 자동으로 채색하는 논문인데, 관심 있으..  (0) 2019.10.31
From Google brain researchers @NeurIPS 2019: Learning to Predict Without Looking Ahead https://www.profillic.com/paper/arxiv:1910.13038 "Rather than hardcoding forward prediction, we try to get agents to *learn* that they need to predict the future"  (0) 2019.10.31
This video gives a quick overview of 41 research papers presented by Google at the International Conference on Computer Vision (ICCV)! [https://youtu.be/z-yvY8iAaHM](https://t.co/1q6od2KUzp?amp=1)  (0) 2019.10.31
From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter ..  (0) 2019.10.29
Turn a line sketch into a photorealistic face: https://www.profillic.com/paper/arxiv:1910.08914 From sparse lines that coarsely describe a face, photorealistic images can be generated using conditional self-attention generative adversarial network (CS..  (0) 2019.10.29
안녕하세요 TmaxData에서 NLP를 연구 중인 장영록입니다:) ALBERT(A Lite BERT - Google 2019.9)라는 논문을 소개드리고자 글 적 습니다. ALBERT는 BERT 보다 모델의 크기는 작지만 GLUE, SQuAD 등의 task에서 더 ..  (0) 2019.10.28
Posted by saveone

댓글을 달아 주세요

안녕하세요, TensorFlow KR 여러분, 저는 현재 AI를 활용한 이미지 변형을 이용해 작업을 제작 중에 있는 미술학도입니다.
TensorFlow KR 여러분들에게 제가 진행중인 작업에 관해 조언을 받고 싶어 이렇게 게시물을 통해 여쭈어봐요.
제가 하고 싶어하는 것은 UGATIT모델을 이용해 제가 갖고 있는 얼굴 이미지들과 제가 변형하여 만든 얼굴 이미지를 학습 시킨 뒤 관객이 자신의 얼굴을 웹캠에 가져다대면 제 형식대로 변형된 모습을 실시간으로 관람할 수 있게 하는 것입니다.
UGATIT 모델은 인물의 사진과 애니메이션 캐릭터 얼굴들을 학습 시킨 뒤, 인물의 사진을 애니메이션화하는데 적용되었던 모델이고 저는 김준호님께서 올려주신 https://github.com/taki0112/UGATIT 코드를 바탕으로 변형하여 진행하고자 해요.
첫 번째 궁금한 사항입니다. Input data를 Webcam에서 특정 시간(예를 들어 2초)을 주기로 받아진 이미지의 변형 된 결과 값만 화면에 보여주고 싶습니다. 혹시 이러한 코드 방식이 가능할지, 참조할 만한 자료 등이나 의견을 주신다면 감사하겠습니다!
두 번째 궁금한 사항은 제가 데스크탑에 연결시킨 웹캠을 이제 구매를 해야하는데, 라즈베리파이로 진행해본 적은 있는데 따로 데스크탑용 웹캠을 사용해본 적이 없거든요, 혹시 웹캠을 이용해 실시간 변환처리를 해보신 분들 중 이거 괜찮았다하는 웹캠이 있을까요? 추천 바랍니다
세 번째 궁금한 사항은, 제 컴퓨터의 GPU는 GTX 1070TI 입니다. 실시간 처리를 진행할 때 1070TI가 혹시 부족할까요? 혹시 GPU 성능이 부족하다고 하면 새로 구매하려고 합니다.

어떠한 의견과 조언도 환영합니다.

감사합니다.
https://www.facebook.com/groups/ReinforcementLearningKR/permalink/2328590640713496/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

#Download #Source #Code from Video Description.
This is complete zero to Amazon Review Sentiment Classification Lesson. In this lesson, I will discuss the following sections
1. What is NLP
2. Applications of NLP
3. Text Data Cleaning Options
4. Bag of Words and TF-IDF (word2vec coming soon)
5. Text Data Preparation
6. Tokenization
7. Lemmatization
8. POS
9. Parsing
10. Named Entity Recognition
11. Text Data Cleaning
12. Model Building
13. Training and Testing
14. Testing with real-world sentences.
Please watch the full video below. Like and Subscribe to show your support.
Sentiment Classification using SpaCy for IMDB and Amazon Review Dataset
https://www.youtube.com/watch?v=cd51nXNpiiU

'Deep Learning > resources' 카테고리의 다른 글

[서버의 VS Code를 띄워놓고 랩탑, 태블릿, 스마트폰에서 코딩하는 방법] 안녕하세요! 지난 겨울 docker로 원격작업하는 방법을 공유했던 연세대학교 통합과정 이응빈입니다. 어제 iPad를 새로 ..  (0) 2019.11.05
머신러닝 모델 디버깅 리소스와 팁 https://t.co/9Y7kDc1hag?amp=1 https://medium.com/infinity-aka-aseem/things-we-wish-we-had-known-before-we-started-our-first-machine-learning-project-336d1d6f2184 https://medium.com/@keeper6928/how-to-uni..  (0) 2019.10.31
#Download #Source #Code from Video Description. This is complete zero to Amazon Review Sentiment Classification Lesson. In this lesson, I will discuss the following sections 1. What is NLP 2. Applications of NLP 3. Text Data Cleaning Options 4. Bag of W..  (0) 2019.10.30
#정보공유 #행사 안녕하세요! RLKorea 운영진입니다! 지난 10월 27~28일 RLKorea Bootcamp가 진행되었는데요! 강화학습의 기초개념인 MDP부터 시작하여 DQN, A2C, DDPG, SAC 등 다양한 강화학습 알고리즘..  (0) 2019.10.29
[XGBoost/LightGBM] Laurae++: xgboost / LightGBM 이번에는 Kaggle에서 가장 많이 사용되는 모델인 xgboost와 lgbm에 관련된 내용입니다. @laurae 님이 만든 xgboost/lightgbm 웹페이지입니다. 공식 documentation에서도 ..  (0) 2019.10.28
안녕하세요 rl kr. 개인적으로 오랜 숙제였던 IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures(IMPALA)를 구현하여 결과를 공유합니다. tensorflow로 구현하였습니다(pytorch로 하지..  (0) 2019.10.28
Posted by saveone

댓글을 달아 주세요

Learn:

1. linear algebra well (e.g. matrix math)
2. calculus to an ok level (not advanced stuff)
3. prob. theory and stats to a good level
4. theoretical computer science basics
5. to code well in Python and ok in C++

Then read and implement ML papers and *play* with stuff! :-)

H / T : Shane Legg

#ArtificialIntelligence101
https://www.facebook.com/groups/ReinforcementLearningKR/permalink/2328590640713496/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

Some of the best courses available on the internet.

1. Data Science Professional Certificate from IBM.
http://bit.ly/2M6o4ko

2. Best Machine Learning course anyone can find on the Internet via Andrew NG/Stanford.
http://bit.ly/2J9uZWV

3. Personal experience, best specialization for Deep Learning and getting advance in Machine Learning via DeeplearningAI.
https://bit.ly/2MjIfd8

4. One the highest rated Machine Learning specialization course with practical approaches and examples via University of Washington.
http://bit.ly/2SuRmKA

5. Introduction to TensorFlow for Artificial Intelligence, Machine Learning, and Deep Learning(4.7 Rating).
http://bit.ly/2M5fLWg
https://m.facebook.com/groups/1376658555701288?view=permalink&id=2840924852607977&sfnsn=mo
Posted by saveone

댓글을 달아 주세요

Deep Learning Drizzle

Drench yourself in Deep Learning, Reinforcement Learning, Machine Learning, Computer Vision, and NLP by learning from these exciting lectures!!

GitHub by Marimuthu Kalimuthu: https://github.com/kmario23/deep-learning-drizzle

Webpage: https://deep-learning-drizzle.github.io

#artificialintelligence #deeplearning #machinelearning #reinforcementlearning
https://www.facebook.com/362056220806309/posts/991017504576841/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video

https://www.profillic.com/paper/arxiv:1910.06407

FireNet: Real-time Segmentation of Fire Perimeter from Aerial Video
https://www.facebook.com/groups/1738168866424224/permalink/2437391486501955/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

From Google brain researchers @NeurIPS 2019: Learning to Predict Without Looking Ahead https://www.profillic.com/paper/arxiv:1910.13038 "Rather than hardcoding forward prediction, we try to get agents to *learn* that they need to predict the future"  (0) 2019.10.31
This video gives a quick overview of 41 research papers presented by Google at the International Conference on Computer Vision (ICCV)! [https://youtu.be/z-yvY8iAaHM](https://t.co/1q6od2KUzp?amp=1)  (0) 2019.10.31
From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter ..  (0) 2019.10.29
Turn a line sketch into a photorealistic face: https://www.profillic.com/paper/arxiv:1910.08914 From sparse lines that coarsely describe a face, photorealistic images can be generated using conditional self-attention generative adversarial network (CS..  (0) 2019.10.29
안녕하세요 TmaxData에서 NLP를 연구 중인 장영록입니다:) ALBERT(A Lite BERT - Google 2019.9)라는 논문을 소개드리고자 글 적 습니다. ALBERT는 BERT 보다 모델의 크기는 작지만 GLUE, SQuAD 등의 task에서 더 ..  (0) 2019.10.28
State of the art in Brain Tumor Segmentation: https://www.profillic.com/paper/arxiv:1909.12901 (A 3D U-net based deep learning model has been trained with the help of brain-wise normalization and patching strategies for the brain tumor segmentation ta..  (0) 2019.10.27
Posted by saveone

댓글을 달아 주세요

#정보공유 #행사

안녕하세요! RLKorea 운영진입니다!

지난 10월 27~28일 RLKorea Bootcamp가 진행되었는데요!

강화학습의 기초개념인 MDP부터 시작하여 DQN, A2C, DDPG, SAC 등 다양한 강화학습 알고리즘과 코드를 살펴보았고 Unity ML-Agents를 이용하여 직접 강화학습 환경을 제작해보기도 했습니다!!

Bootcamp에서 진행한 강의자료와 코드들을 저장한 Github Repository를 공유드립니다! 링크는 다음과 같습니다.

[https://github.com/reinforcement-learning-kr/rl_bootcamp](https://github.com/reinforcement-learning-kr/rl_bootcamp)

또한 행사 사진도 함께 공유합니다!

정말 멋진 장소를 후원해주신 마이크로소프트 코리아와 행사에 참여해주신 모든 분들께 진심으로 감사드립니다! :)
https://www.facebook.com/groups/ReinforcementLearningKR/permalink/2328590640713496/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

"Prior가 뭔지는 알잖아요?"

동의하는척 웃고 넘어간 뒤 한 없이 부끄러워져서 머신러닝 관련 기본적인 통계추론 내용을 정리를 해보았습니다. 논문 읽을 때 자주 나오는 내용인데도 생각보다 간결히 정리하기가 어려웠습니다. 대단한 내용이 있는 건 아니지만 잘 대답할 수 있나 확인해보시면 좋겠습니다.

.
.
.

1. 머신러닝 할 때 통계가 왜 필요할까?

2. PDF의 값이 1이 넘을 수 있겠는가? 예시를 들자면?

3. 아는 확률분포와 언제 사용할 수 있는지 말해보라.

4. 중심극한정리가 무엇이고 어느 맥락에서 사용되는지 설명해보아라.

5. 모수적 방법과 비모수적 방법의 차이는? 모수적 방법을 사용할 수 없는 경우는 언제인가?

6. 우도(likelihood)를 설명해 보아라.

7. 키가 측정할 때 마다 다르다면 어떻게 참값을 추정할 수 있을까?(키가 더이상 크지 않는 상황을 가정하자 ㅠㅠ)

8. MLE와 MAP차이를 수식으로 쓸 수 있겠는가? 그 의미는? 각 방법의 한계점은 뭐가 있을까? 사전확률(prior)을 설명해 보아라.

9. MLE 방법과 Mean Squared Error, Cross Entropy를 줄이는 방법이 동치인 경우를 말해보아라.

10. Log Likelihood를 쓰는 이유는 뭘까? log를 씌우면 MLE의 값이 변하지 않을까?

11. pure, semi, naive bayesian의 차이가 뭘까? 어떤 장단점이 있을까?

12. (NLP) MLE와 naive-bayesian을 사용한 Unigram Language Modeling의 학습 방법을 설명해보아라. 이 방법의 문제점은 뭘까?

블로그에 간략히 답을 달았습니다.
https://www.facebook.com/groups/modulabs/permalink/2541750722556762/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

Check out these Top  #MachineLearning Youtube Videos Under 10 Minutes
https://www.facebook.com/groups/aiIDEASandINNOVATIONS/permalink/2847112431989219/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

Turn a line sketch into a photorealistic face:
https://www.profillic.com/paper/arxiv:1910.08914

From sparse lines that coarsely describe a face, photorealistic images can be generated using conditional self-attention generative adversarial network (CSAGAN)

"LinesToFacePhoto: Face Photo Generation from Lines with Conditional Self-Attention Generative Adversarial Network"
https://www.facebook.com/groups/ReinforcementLearningKR/permalink/2324753411097219/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

This video gives a quick overview of 41 research papers presented by Google at the International Conference on Computer Vision (ICCV)! [https://youtu.be/z-yvY8iAaHM](https://t.co/1q6od2KUzp?amp=1)  (0) 2019.10.31
From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter ..  (0) 2019.10.29
Turn a line sketch into a photorealistic face: https://www.profillic.com/paper/arxiv:1910.08914 From sparse lines that coarsely describe a face, photorealistic images can be generated using conditional self-attention generative adversarial network (CS..  (0) 2019.10.29
안녕하세요 TmaxData에서 NLP를 연구 중인 장영록입니다:) ALBERT(A Lite BERT - Google 2019.9)라는 논문을 소개드리고자 글 적 습니다. ALBERT는 BERT 보다 모델의 크기는 작지만 GLUE, SQuAD 등의 task에서 더 ..  (0) 2019.10.28
State of the art in Brain Tumor Segmentation: https://www.profillic.com/paper/arxiv:1909.12901 (A 3D U-net based deep learning model has been trained with the help of brain-wise normalization and patching strategies for the brain tumor segmentation ta..  (0) 2019.10.27
ICYMI from BMVC 2019: human motion transfer - generation of a video https://www.profillic.com/paper/arxiv:1910.09139 (Their GAN-based architecture, DwNet, leverages dense intermediate pose-guided representation and refinement process to warp the requi..  (0) 2019.10.26
Posted by saveone

댓글을 달아 주세요

[XGBoost/LightGBM] Laurae++: xgboost / LightGBM

이번에는 Kaggle에서 가장 많이 사용되는 모델인 xgboost와 lgbm에 관련된 내용입니다.

@laurae 님이 만든 xgboost/lightgbm 웹페이지입니다. 공식 documentation에서도 링크를 제공하고 있습니다.

xgboost와 lightgbm의 parameter에 대한 설명들을 볼 수 있습니다.
총 86개의 parameter에 대한 다음과 같은 내용이 정리되어 있고, 원하는 filter로 parameter를 선택해서 볼 수도 있습니다.

- 간단 설명
- 매개변수의 유형과 카테고리
- xgboost와 lightgbm에서의 명칭
- 범위
- major impact와 minor impact
- 일반적인 사용법과 tips
- 그 외 세부적인 설명과 유용한 링크

링크는 다음과 같습니다.

https://sites.google.com/view/lauraepp/parameters

사이트에는 튜토리얼과 벤치마크 결과도 있으니 참고하면 좋을 것 같습니다. 다들 즐거운 Kaggle/ML 생활하세요 :)

+ 원래는 한글로 번역을 할까 싶었는데, 번역본보다 원문이 이해가 잘되서 포기했습니다.
+ 이 표는 awesome table이라는 서비스로 만들었다는데, 꽤 괜찮아보입니다.

#Kaggle #LightGBM #XGBoost #parameter #Laurae
https://www.facebook.com/groups/KaggleKoreaOpenGroup/permalink/515068679225236/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

DEVIEW2019 Keynote에서 “석상옥 대표님”이 소개해주신 NAVER LABS의 자율주행용 Open dataset입니다. 국내자율주행 기술 성장에 큰 도움이 될 것으로 기대됩니다 : )
https://www.facebook.com/groups/ReinforcementLearningKR/permalink/2324753411097219/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

안녕하세요 TmaxData에서 NLP를 연구 중인 장영록입니다:)

ALBERT(A Lite BERT - Google 2019.9)라는 논문을 소개드리고자 글 적 습니다.

ALBERT는 BERT 보다 모델의 크기는 작지만 GLUE, SQuAD 등의 task에서 더 높은 성능을 달성한 모델입니다. Downstream Task에 높은 성능을 얻은 것도 중요하지만 Transformer의 각 Layer 간 Parameter를 공유하여 모델의 크기가 BERT 보다 현저히 줄었다는게 가장 큰 Contribution인 것 같습니다.

논문 내용을 정리한 제 블로그 글을 공유드리니 관심있으신 분은 보시길 바랍니다. :)

논문 링크 : [https://arxiv.org/abs/1909.11942](https://arxiv.org/abs/1909.11942)

논문 정리 블로그 : [https://y-rok.github.io/nlp/2019/10/23/albert.html](https://y-rok.github.io/nlp/2019/10/23/albert.html)
https://www.facebook.com/groups/TensorFlowKR/permalink/1020923228248735/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

From NeurIPS 2019: Particularly helpful for fighting wild forest fires: real-time segmentation of fire perimeter from aerial full-motion infrared video https://www.profillic.com/paper/arxiv:1910.06407 FireNet: Real-time Segmentation of Fire Perimeter ..  (0) 2019.10.29
Turn a line sketch into a photorealistic face: https://www.profillic.com/paper/arxiv:1910.08914 From sparse lines that coarsely describe a face, photorealistic images can be generated using conditional self-attention generative adversarial network (CS..  (0) 2019.10.29
안녕하세요 TmaxData에서 NLP를 연구 중인 장영록입니다:) ALBERT(A Lite BERT - Google 2019.9)라는 논문을 소개드리고자 글 적 습니다. ALBERT는 BERT 보다 모델의 크기는 작지만 GLUE, SQuAD 등의 task에서 더 ..  (0) 2019.10.28
State of the art in Brain Tumor Segmentation: https://www.profillic.com/paper/arxiv:1909.12901 (A 3D U-net based deep learning model has been trained with the help of brain-wise normalization and patching strategies for the brain tumor segmentation ta..  (0) 2019.10.27
ICYMI from BMVC 2019: human motion transfer - generation of a video https://www.profillic.com/paper/arxiv:1910.09139 (Their GAN-based architecture, DwNet, leverages dense intermediate pose-guided representation and refinement process to warp the requi..  (0) 2019.10.26
From ICCV: Gaze360, a large-scale gaze-tracking dataset and method for robust 3D gaze estimation in unconstrained images https://www.profillic.com/paper/arxiv:1910.10088  (0) 2019.10.26
Posted by saveone

댓글을 달아 주세요

안녕하세요 rl kr. 개인적으로 오랜 숙제였던 IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures(IMPALA)를 구현하여 결과를 공유합니다. tensorflow로 구현하였습니다(pytorch로 하지 못해 토치 유저분들에게는 죄송하다는 말씀을 올리며). 사용한 것은 distributed tensorflow를 기본적으로 사용하였습니다. 제 예전 actor critic으로 breakout을 잘 배우기 위해서는 엄청나게 오랜시간(10시간정도)걸렸지만 20개의 actor로 2시간만에 의미있는 결과를 뽑아낼 수 있었습니다. 혹시 코드에서 오류 혹은 수식을 코드로 옮기는 과정에서 잘못된 부분이 있다면 바로 알려주시면 감사하겠습니다.

ps. 윤수로님께 감사하다는 말씀 올립니다.
https://www.facebook.com/groups/ReinforcementLearningKR/permalink/2324753411097219/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

In this lesson, we will discuss how you can expand the NLP model in SpaCy. Text data is so wide therefore it is possible that the model is not present in Spacy to deal with it and in that case you can build your own rule and merge with the default SpaCy model.

NLP Tutorial 7 - Combining NLP Models and Custom Rules in SpaCy Python Tutorial
https://youtu.be/yHmfOWryK4M
https://m.facebook.com/groups/107107546348803?view=permalink&id=984324795293736&sfnsn=mo

'Deep Learning > course' 카테고리의 다른 글

Deep Learning Drizzle Drench yourself in Deep Learning, Reinforcement Learning, Machine Learning, Computer Vision, and NLP by learning from these exciting lectures!! GitHub by Marimuthu Kalimuthu: https://github.com/kmario23/deep-learning-drizzle We..  (0) 2019.10.29
Check out these Top #MachineLearning Youtube Videos Under 10 Minutes  (0) 2019.10.29
In this lesson, we will discuss how you can expand the NLP model in SpaCy. Text data is so wide therefore it is possible that the model is not present in Spacy to deal with it and in that case you can build your own rule and merge with the default SpaCy..  (0) 2019.10.27
#Download working file from the video description Processing Pipeline is an important step in feature extraction from text data. Learn how the pipeline works and how you can speed up the processing by disabling some blocks of the pipeline. NLP Tutorial ..  (0) 2019.10.27
Deep Learning with Keras A-Z: Part of our Data Science & Machine Learning Notes Series. We will now explore step by step tutor using Keras on MNIST. Go through the Analysis steps and try to understand the context. If some topics are too advanced they wi..  (0) 2019.10.27
버클리대학에서 진행한 Introduction to Deep Learning (STAT 157)강의의 강의 슬라이드와 강의 영상입니다. 한글화된 Dive into Deep Learning(https://ko.d2l.ai/)과 함께 보시면 처음 공부하시는 분들께 많은 도움..  (0) 2019.10.25
Posted by saveone

댓글을 달아 주세요

#Download working file from the video description
Processing Pipeline is an important step in feature extraction from text data. Learn how the pipeline works and how you can speed up the processing by disabling some blocks of the pipeline.
NLP Tutorial 8 - Introduction to Processing Pipeline in SpaCy Python Tutorial | SpaCy for NLP
https://youtu.be/GI_2wmSK49I
https://m.facebook.com/groups/107107546348803?view=permalink&id=984991961893686&sfnsn=mo
Posted by saveone

댓글을 달아 주세요

State of the art in Brain Tumor Segmentation:
https://www.profillic.com/paper/arxiv:1909.12901

(A 3D U-net based deep learning model has been trained with the help of brain-wise normalization and patching strategies for the brain tumor segmentation task)
https://www.facebook.com/groups/DeepNetGroup/permalink/985359465190269/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

Turn a line sketch into a photorealistic face: https://www.profillic.com/paper/arxiv:1910.08914 From sparse lines that coarsely describe a face, photorealistic images can be generated using conditional self-attention generative adversarial network (CS..  (0) 2019.10.29
안녕하세요 TmaxData에서 NLP를 연구 중인 장영록입니다:) ALBERT(A Lite BERT - Google 2019.9)라는 논문을 소개드리고자 글 적 습니다. ALBERT는 BERT 보다 모델의 크기는 작지만 GLUE, SQuAD 등의 task에서 더 ..  (0) 2019.10.28
State of the art in Brain Tumor Segmentation: https://www.profillic.com/paper/arxiv:1909.12901 (A 3D U-net based deep learning model has been trained with the help of brain-wise normalization and patching strategies for the brain tumor segmentation ta..  (0) 2019.10.27
ICYMI from BMVC 2019: human motion transfer - generation of a video https://www.profillic.com/paper/arxiv:1910.09139 (Their GAN-based architecture, DwNet, leverages dense intermediate pose-guided representation and refinement process to warp the requi..  (0) 2019.10.26
From ICCV: Gaze360, a large-scale gaze-tracking dataset and method for robust 3D gaze estimation in unconstrained images https://www.profillic.com/paper/arxiv:1910.10088  (0) 2019.10.26
State of the art in object detection: FuseMODNet: Real-Time Camera and LiDAR based Moving Object Detection for robust low-light Autonomous Driving https://www.profillic.com/paper/arxiv:1910.05395  (0) 2019.10.25
Posted by saveone

댓글을 달아 주세요

정말 대단함!

일독을 권합니다~
https://www.facebook.com/159255300951476/posts/1172071929669803/?sfnsn=mo

'StartUp(창업) > 스타트업 조언' 카테고리의 다른 글

정말 대단함! 일독을 권합니다~  (0) 2019.10.27
Posted by saveone

댓글을 달아 주세요

Deep Learning with Keras A-Z: Part of our Data Science & Machine Learning Notes Series. We will now explore step by step tutor using Keras on MNIST. Go through the Analysis steps and try to understand the context. If some topics are too advanced they will be covered soon in the upcoming Learning Notes Series.

Get your Notes here:[http://bit.ly/2MQVcgf](https://l.facebook.com/l.php?u=https%3A%2F%2Fbit.ly%2F2MQVcgf%3Ffbclid%3DIwAR0XWCuw5ALJ7YDyvPaED-6nFeM8e_upLhqUjmqS9TV6QGgfhZU1UUbd_Jc&h=AT2-kOSNPPdv1ICdwKZbI3_Je7jTjOfa1KYaQYdZsuDgIf-NxK_J9lEuYAfhWdyrkh72wv3uDMn65K7WHMrGPdH9aPjkXCWUP4mYIIwD3948AV5lZ6sSkUDY8R9iUj_e1R9P4CIfcapKfTwP79NT1qhrdZBXt2Z7kaC1AMnvb4tFLMGdZqWWRgWmKPVM4gTtuFThQEceooKOAVRqZh3DSUp2TAGIOv4kApQ824OQ3ftbW3HvOCrVSxU6TuiX65AXdFLXihO5MxThizVIsf6h-8HP1oOYFUvarHZjB2kteZNE9F7OTTKuy4oIfDkEPUABgkGnjlEoExlVfARkHSIb6U1POYGvL1DxQKvug-F3AwFmd0VW3V2kaXHPKJl_HQvIDhr_bYaP3j0ZAjxaj6eZcfiaolUvW8h7Je2PEy6VXhhYNrJePd9GMA0Tksh74rSV7_kDy_42tJ8d7VzVg969nLFqAldvkfJbY56RmOqLMfVNweGyrRYBx4MGjUTusjl5qJ4KO4GtxWZZg3W0iGF1pSNnVduzHePYWxLofwrinS94Ya4ZHrVMIv0EoeA4_lfSAqCyugMGBVNAgXLFfBgtIoYkqyCTWd8Z5EVXhK1Z2CdxsrhwQI6GdO7jLyCbXj0QkM0ydJk)
https://www.facebook.com/groups/DataScienceGroup/permalink/2731937136868159/?sfnsn=mo
Posted by saveone

댓글을 달아 주세요

ICYMI from BMVC 2019: human motion transfer - generation of a video
https://www.profillic.com/paper/arxiv:1910.09139

(Their GAN-based architecture, DwNet, leverages dense intermediate pose-guided representation and refinement process to warp the required subject appearance, in the form of the texture, from a source image into a desired pose.)
https://www.facebook.com/groups/DeepNetGroup/permalink/984436071949275/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

안녕하세요 TmaxData에서 NLP를 연구 중인 장영록입니다:) ALBERT(A Lite BERT - Google 2019.9)라는 논문을 소개드리고자 글 적 습니다. ALBERT는 BERT 보다 모델의 크기는 작지만 GLUE, SQuAD 등의 task에서 더 ..  (0) 2019.10.28
State of the art in Brain Tumor Segmentation: https://www.profillic.com/paper/arxiv:1909.12901 (A 3D U-net based deep learning model has been trained with the help of brain-wise normalization and patching strategies for the brain tumor segmentation ta..  (0) 2019.10.27
ICYMI from BMVC 2019: human motion transfer - generation of a video https://www.profillic.com/paper/arxiv:1910.09139 (Their GAN-based architecture, DwNet, leverages dense intermediate pose-guided representation and refinement process to warp the requi..  (0) 2019.10.26
From ICCV: Gaze360, a large-scale gaze-tracking dataset and method for robust 3D gaze estimation in unconstrained images https://www.profillic.com/paper/arxiv:1910.10088  (0) 2019.10.26
State of the art in object detection: FuseMODNet: Real-Time Camera and LiDAR based Moving Object Detection for robust low-light Autonomous Driving https://www.profillic.com/paper/arxiv:1910.05395  (0) 2019.10.25
안녕하세요. 각기 다른 Receptive Field 를 가진 컨볼루션 필터로부터 출력되는 피쳐맵 간에 적응적인 Weighted Average 연산을 통해 작업(Image classification) 성능을 끌어올릴 수 있는 어텐션 모듈을 제안..  (0) 2019.10.24
Posted by saveone

댓글을 달아 주세요

From ICCV: Gaze360, a large-scale gaze-tracking dataset and method for robust 3D gaze estimation in unconstrained images

https://www.profillic.com/paper/arxiv:1910.10088
https://www.facebook.com/groups/TensorFlowKR/permalink/1018870715120653/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

State of the art in Brain Tumor Segmentation: https://www.profillic.com/paper/arxiv:1909.12901 (A 3D U-net based deep learning model has been trained with the help of brain-wise normalization and patching strategies for the brain tumor segmentation ta..  (0) 2019.10.27
ICYMI from BMVC 2019: human motion transfer - generation of a video https://www.profillic.com/paper/arxiv:1910.09139 (Their GAN-based architecture, DwNet, leverages dense intermediate pose-guided representation and refinement process to warp the requi..  (0) 2019.10.26
From ICCV: Gaze360, a large-scale gaze-tracking dataset and method for robust 3D gaze estimation in unconstrained images https://www.profillic.com/paper/arxiv:1910.10088  (0) 2019.10.26
State of the art in object detection: FuseMODNet: Real-Time Camera and LiDAR based Moving Object Detection for robust low-light Autonomous Driving https://www.profillic.com/paper/arxiv:1910.05395  (0) 2019.10.25
안녕하세요. 각기 다른 Receptive Field 를 가진 컨볼루션 필터로부터 출력되는 피쳐맵 간에 적응적인 Weighted Average 연산을 통해 작업(Image classification) 성능을 끌어올릴 수 있는 어텐션 모듈을 제안..  (0) 2019.10.24
ICYMI: Faceswap+ "Mask-Guided Portrait Editing with Conditional GANs" https://www.profillic.com/paper/arxiv:1905.10346 They address three issues in existing techniques: diversity, quality, and controllability for portrait synthesis and editing.  (0) 2019.10.24
Posted by saveone

댓글을 달아 주세요

State of the art in object detection: FuseMODNet: Real-Time Camera and LiDAR based Moving Object Detection for robust low-light Autonomous Driving

https://www.profillic.com/paper/arxiv:1910.05395
https://www.facebook.com/groups/1738168866424224/permalink/2431434923764278/?sfnsn=mo

'Deep Learning > Papers2read' 카테고리의 다른 글

ICYMI from BMVC 2019: human motion transfer - generation of a video https://www.profillic.com/paper/arxiv:1910.09139 (Their GAN-based architecture, DwNet, leverages dense intermediate pose-guided representation and refinement process to warp the requi..  (0) 2019.10.26
From ICCV: Gaze360, a large-scale gaze-tracking dataset and method for robust 3D gaze estimation in unconstrained images https://www.profillic.com/paper/arxiv:1910.10088  (0) 2019.10.26
State of the art in object detection: FuseMODNet: Real-Time Camera and LiDAR based Moving Object Detection for robust low-light Autonomous Driving https://www.profillic.com/paper/arxiv:1910.05395  (0) 2019.10.25
안녕하세요. 각기 다른 Receptive Field 를 가진 컨볼루션 필터로부터 출력되는 피쳐맵 간에 적응적인 Weighted Average 연산을 통해 작업(Image classification) 성능을 끌어올릴 수 있는 어텐션 모듈을 제안..  (0) 2019.10.24
ICYMI: Faceswap+ "Mask-Guided Portrait Editing with Conditional GANs" https://www.profillic.com/paper/arxiv:1905.10346 They address three issues in existing techniques: diversity, quality, and controllability for portrait synthesis and editing.  (0) 2019.10.24
안녕하세요 ! 운영하고 있는 딥러닝논문읽기모임의 열 네번째 유튜브 영상이 업로드 되어 공유합니다.논문은 자연어 처리의 절대강자 Bert 계열 논문인 ‘RoBERTa’ 입니다 ! 자연어처리 팀 이신..  (0) 2019.10.21
Posted by saveone

댓글을 달아 주세요