Diese Präsentation wurde erfolgreich gemeldet.
Wir verwenden Ihre LinkedIn Profilangaben und Informationen zu Ihren Aktivitäten, um Anzeigen zu personalisieren und Ihnen relevantere Inhalte anzuzeigen. Sie können Ihre Anzeigeneinstellungen jederzeit ändern.

【14-C-7】コンピュータビジョンを支える深層学習技術の新潮流

Developers Summit 2019【14-C-7】鮫島様の講演資料です。

  • Als Erste(r) kommentieren

【14-C-7】コンピュータビジョンを支える深層学習技術の新潮流

  1. 1. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Masaki Samejima Machine Learning Solutions Architect, Amazon Web Services Japan. 2019.2.14 Developers Summit 2019
  2. 2. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Agenda • • • •
  3. 3. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  4. 4. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • Demographic Data Facial Landmarks Sentiment Expressed Image Quality General Attributes
  5. 5. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 2012 SuperVision[1] ILSVRC2012 [1] A. Krizhevsky, et al., Imagenet classification with deep convolutional neural networks, NIPS 2012. [2] R Girshick, et al., Rich feature hierarchies for accurate object detection and semantic segmentation, CVPR 2014. [3] I.J. Goodfellow, et al., Generative Adversarial Nets, NIPS 2014. [4] V. Badrinarayanan, et al, SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. PAMI 2017 2014 R-CNN[2] Pascal VOC GAN[3] SegNet[4] 2015
  6. 6. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. https://gluon-cv.mxnet.io/model_zoo/classification.html senet_154 resnet_v1d resnet_v1c resnet_v1b resnet_v1 densenet darknet VGG resnet_v2 mobilenet mobilenetv2 0.80 0.75 0.70 Accuracy 1000 2000 #sample/sec.3000 4000 • ImageNet 80% • V100 GPU
  7. 7. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. https://gluon-cv.mxnet.io/model_zoo/detection.html mAP 10 100 #sample/sec. 40 35 30 yolo3 faster_rcnn ssd • (IoU ) mAP 30-40% •
  8. 8. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. https://gluon-cv.mxnet.io/model_zoo/segmentation.html 0 10 20 30 40 50 60 70 80 90 100 fcn_resnet101 psp_resnet101 deeplab_resnet101 fcn_resnet101 psp_resnet101 deeplab_resnet101 deeplab_resnet152 COCO VOC IoU
  9. 9. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 3 [1] [1] B. Tekin, et al., Real-Time Seamless Single Shot 6D Object Pose Prediction, CVPR 2018. [2] R. Girdhar, et al., Detect-and-Track: Efficient Pose Estimation in Videos, CVPR 2018. [3] L. Chen, et al., MaskLab: Instance Segmentation by Refining Object Detection with Semantic and Direction Features, CVPR 2018. [2] [3]
  10. 10. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. GANNoise Text-to-image [3] (and Image-to-text)[1] [2] [1] P. Isola, et al., Image-to-Image Translation with Conditional Adversarial Nets, CVPR 2017. [2] C. Ledig, et al., Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network, CVPR 2017. [3] S. Reed, et al., Generative Adversarial Text to Image Synthesis, ICML 2016.
  11. 11. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Saliency ( ) [1] [1] N. Liu, et al., PiCANet: Learning Pixel-wise Contextual Attention for Saliency Detection, CVPR 2018. [2] Z. Li, et al., MegaDepth: Learning Single-View Depth Prediction from Internet Photos, CVPR 2018. [2]
  12. 12. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 0 2 4 6 8 10 12 14 16 18 20 1 2 3 4 5 6 7 8 9 1011121314151617181920 ID [1] O. Vinyals, et al., Matching Networks for One Shot Learning, arXiv:1606.04080 • • [1]
  13. 13. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • Deep Learning • X. Yuan, et al., Adversarial Examples: Attacks and Defenses for Deep Learning, IEEE Trans Neural Netw Learn Syst. 2019.
  14. 14. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  15. 15. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • • • ONNX AutoML Define-by-run
  16. 16. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • TensorFlow models TF slim GluonCV ChainerCV PyTorchCV
  17. 17. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. ResNet (Gluon vs MXNet) num_unit = len(units) assert(num_unit == num_stages) data = mx.sym.Variable(name='data') if dtype == 'float32': data = mx.sym.identity(data=data, name='id') else: if dtype == 'float16': data = mx.sym.Cast(data=data, dtype=np.float16) data = mx.sym.BatchNorm(data=data, fix_gamma=True, eps=2e-5, momentum=bn_mom, name='bn_data') (nchannel, height, width) = image_shape if height <= 32: # such as cifar10 body = mx.sym.Convolution(data=data, num_filter=filter_list[0], kernel=(3, 3), stride=(1,1), pad=(1, 1), no_bias=True, name="conv0", workspace=workspace) else: # often expected to be 224 such as imagenet body = mx.sym.Convolution(data=data, num_filter=filter_list[0], kernel=(7, 7), stride=(2,2), pad=(3, 3), no_bias=True, name="conv0", workspace=workspace) body = mx.sym.BatchNorm(data=body, fix_gamma=False, eps=2e-5, momentum=bn_mom, name='bn0') body = mx.sym.Activation(data=body, act_type='relu', name='relu0') body = mx.sym.Pooling(data=body, kernel=(3, 3), stride=(2,2), pad=(1,1), pool_type='max') for i in range(num_stages): body = residual_unit(body, filter_list[i+1], (1 if i==0 else 2, 1 if i==0 else 2), False, name='stage%d_unit%d' % (i + 1, 1), bottle_neck=bottle_neck, workspace=workspace, memonger=memonger) for j in range(units[i]-1): body = residual_unit(body, filter_list[i+1], (1,1), True, name='stage%d_unit%d' % (i + 1, j + 2), bottle_neck=bottle_neck, workspace=workspace, memonger=memonger) bn1 = mx.sym.BatchNorm(data=body, fix_gamma=False, eps=2e-5, momentum=bn_mom, name='bn1') relu1 = mx.sym.Activation(data=bn1, act_type='relu', name='relu1') MXNet from mxnet.gluon.model_zoo import vision resnet18 = vision.resnet18_v1() Gluon
  18. 18. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. ONNX (Open Neural Network Exchange) MXNet Caffe2 PyTorch TF CNTKCoreML Tensor RT NGraph SNPE • ONNX ONNX •
  19. 19. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. ONNX Protocol Buffers • • • API Protocol Buffers Graph Operator Tensor, … Operator Definitions ONNX Python API
  20. 20. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Define-and-run Define-by-run • Define-and-run • • TensorFlow, MXNet • Define-by-run • • Chainer PyTorch, TensorFlow, MXNet
  21. 21. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Define-and-run Define-by-run Define-and-run Define-by-run def our_function(A, B): C = A + B return C A = Load_Data_A() B = Load_Data_B() result = our_function(A, B) A = placeholder() B = placeholder() C = A + B our_function = compile(inputs=[A, B], outputs =[C]) A = Load_Data_A() B = Load_Data_B() result = our_function(A, B) https://gluon.mxnet.io/chapter07_distributed-learning/hybridize.html Define Run Define, Run
  22. 22. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Define-by-run Define-and-run Define-by-run def our_function(A, B): C = A + B return C A = Load_Data_A() B = Load_Data_B() result = our_function(A, B) A = placeholder() B = placeholder() C = A + B our_function = compile(inputs=[A, B], outputs =[C]) A = Load_Data_A() B = Load_Data_B() result = our_function(A, B)
  23. 23. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AutoML • • , etc. D. Bayor, et al., TFX: A TensorFlow-Based Production-Scale Machine Learning Platform, KDD 2017.
  24. 24. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AutoML • AutoML • ICML 2014 AutoML * • • • Meta-Learning, Learning to learn * https://sites.google.com/site/automlwsicml14/
  25. 25. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AutoML
  26. 26. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. AutoML Amazon Forecast User CSV file 1. S3 2. Forecast 3. Forecast 4.
  27. 27. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  28. 28. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • Model Server Interpretable ML
  29. 29. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Model Server • • Model Server • • REST/RPC Model Server Mobile client Deploy REST/RPC
  30. 30. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. TensorFlow Serving [1] C. Olston, et al., TensorFlow-Serving: Flexible, High-Performance ML Serving, NIPS 2017. • Controller, Synchronizer Serving job • Router Serving job
  31. 31. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. MXNet Model Server https://aws.amazon.com/jp/blogs/news/model-server-for-apache-mxnet-v1-0-released/ • REST API • MMS 1.0 1,000 MMS 1.0 MMS 0.4
  32. 32. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • • • AWS, SageMaker Neo • Nvidia, TensorRT Raspberry Pi ResNet18 Mobilenet 11.5x 2.2x
  33. 33. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. SageMaker Neo / TVM • Operator Fusion • Data Layout Transformation 4x4 4x4 • Tensor Expression and Schedule Space • Nested Parallelism with Cooperation • etc… T. Chen, et al., TVM: An Automated End-to-End Optimizing Compiler for Deep Learning, OSDI 2018.
  34. 34. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. TensorRT • Layer & Tensor Fusion 1 • FP16 and INT8 Precision Calibration FP32 FP16 INT8 • Kernel Auto-Tuning • Dynamic Tensor Memory • Multi Stream Execution https://devblogs.nvidia.com/tensorrt-3-faster-tensorflow-inference/
  35. 35. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Interpretable ML: : SVM GBT C. Molnar, Interpretable Machine Learning, https://christophm.github.io/interpretable-ml-book/ >900< 900 < 2000 km2 > 2000 km2
  36. 36. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Interpretable ML for computer vision • • M.T. Ribeiro, et al., Anchors: High-Precision Model-Agnostic Explanations, AAAI 2018.
  37. 37. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  38. 38. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • • • 1 1 • • AWS Inferentia • Intel Nervana
  39. 39. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Machine Learning on FPGA • FPGA • AWS F1 instance Amazon Machine Image • Loop tiling [1] [1] C. Zhang, et al., Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks, FPGA 2015.
  40. 40. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • GPU
  41. 41. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  42. 42. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  43. 43. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. • • AutoML AI •
  44. 44. © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. https://amzn.to/aws_dev

×