SlideShare a Scribd company logo
1 of 42
小數據如何實現電腦視覺
賴尚宏
1
Principal Researcher
Microsoft AI R&D Center
Taipei
Professor
Dept. of Computer Science
National Tsing Hua University
DEEP
LEARNING
→
COMPUTER
VISION
➢Success in applying deep learning to a
number of computer vision problems
➢Image classification
➢Image segmentation
➢Object detection, tracking, recognition
➢Image processing (style transfer, super-resolution,
deblurring, etc.)
➢3D reconstruction
➢Image/video captioning/Q&A
➢…….
2
CNN (CONVOLUTIONAL NEURAL NETWORKS)
CNN has been successfully used in many image classification and object
detection problems.
The most famous CNN architecture is LeNet used to classify hand-written digits.
Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner. Gradient-based learning applied to document recognition. Proc. of IEEE, 1998
MODEL COMPLEXITY
The complexity of a learning model can be characterized by the number of
parameters in the model.
A large number of model parameters means it can learn/model very complex
structure.
Recent deep networks usually contain millions of parameters to be learned from
the training data.
4Figures adopted from Deep Learning, by I. Goodfellow, Y. Bengio and A. Courville, MIT Press
DATASET SIZE
5
The age of “Big Data” has made machine learning much easier because it provides a
very large number of training data of wide varieties for the deep networks (millions of
parameters) to learn to achieve the level of human capability.
Figures adopted from Deep Learning, by I. Goodfellow, Y. Bengio and A. Courville, MIT Press
FACE RECOGNITION DATASET
Dataset Subjects Images Note
CASIA-WebFace 10,575 494,414 The second largest public dataset available for
face verification and recognition problems.
VGGFace2 9131 3.31 million Images have large variations in pose, age,
illumination, ethnicity and profession.
MS-Celeb-1M 100k 10 million To facilitate the IRC face recognition task
LFW 5749 13,233 For face verification
FaceScrub 530 107,818 One of the largest public face databases.
MegaFace 672,000+ 4.7 million the largest (in number of identities) publicly
available facial recognition dataset
CelebA 10,177 202,599 5 landmark locations, 40 binary attributes
annotations per image.
MultiPie 337 750,000+ Under 15 view points and 19 illumination
conditions
UMDFaces 8,277 367,888 Contains both still images and video frames.
DATA ANNOTATION
➢Image-level annotation
➢Object-level annotation
7
{ edge, center } { edge (point),
center (point) }
{ edge (b-box),
center (b-box) }
{ edge (pixel labels),
center (pixel labels) }
very fast fast slow very slow
PROBLEMS WITH PREPARING TRAINING DATA
➢Require large amount of training data of wide varieties
➢Imbalanced data
➢Usually domain specific
➢Intensive human labeling cost
➢Restricted to supervised learning
8
DATA AUGMENTATION
9
DATA
AUGMENTATION
State-of-the-art neural networks typically have parameters in
the order of millions. The number of parameters needed
is proportional to the complexity of the task your model has
to perform.
The size of data samples for training the model should be
proportional to the number of parameters of the model.
10
COMMON DATA
AUGMENTATION METHODS
➢Flipping
➢Rotation
➢Scaling
➢Cropping
➢Translation
➢Adding Noise
➢Color Transformation
11
Data Augmentation for Training Detector of Partially
Occluded Cars
Augmentation strategy:
resize the vehicles in the training
images with 0.1 to 0.9 of the
original size
horizontally flip the images with
0.5 probability.
truncate a vehicle with 0.5
probability by a new window such
that the vehicle contains at least
25% of its appearance visible.
12
TRANSFER LEARNING
13
TRANSFER
LEARNING
14
Transfer learning is a machine learning
method where a model trained for a task is
reused as the starting point for re-training
a model for another task.
It’s popular for some computer vision tasks
that require vast compute and time
resources to train large and deep neural
network models.
This learning process will tend to work if the
features are general, meaning suitable to
both base and target tasks, instead of
specific to the base task.
TRANSFER LEARNING WITH CNNS
15Slide credit: Stanford Univ. CS 231n
GENERAL GUIDELINE FOR FINETUNING DEEP NEURAL
NETWORKS
16Slide credit: Stanford Univ. CS 231n
SEMI-SUPERVISED LEARNING
17
INTRODUCTION
Supervised Learning – Train a model with labeled data only.
Semi-Supervised Learning – Train a model with labeled(R) and
unlabeled(U) data, usually U>>R.
Why semi-supervised learning ?
Collecting data is easy, but collecting labelled data is expensive.
18
Some Labeled Data
Lots of Unlabeled Data
Model
Semi-supervised
Learning
PSEUDO LABELING
19
Example:
Cat or Dog Classification
DOG
CAT
Labeled Data
Unlabeled Data
1/5
DOG
CAT
Labeled Data
2/5
TRAIN
Unlabeled Data 3/5
PREDICTPREDICT
DOG
CAT
4/5
TRAIN
5/5
DISCUSSION
Collecting labelled data is expensive
More accurate decision boundary with
unlabeled instances
20
Decision
boundary
Labeled
Instances
Decision
boundary
Unlabeled
Instances
• No way to verify the produced labels’
accuracy
→ less trust worthy
Pros Cons
GAN FOR IMAGE SYNTHESIS
(Generative Adversarial Network)
21
GAN ARCHITECTURE
22
Real
or
Fake
Loss
G
Generator Network
D
Discriminator Network
Fake Images G(z)
Real Images x
min
𝜃 𝑔
max
𝜃 𝑑
ቈ
቉
𝔼 𝑥∼𝑝 𝑑𝑎𝑡𝑎
log 𝐷 𝜃 𝑑
𝑥 + 𝔼 𝑧∼𝑝 𝑧
log ቆ
ቇ
1
− 𝐷 𝜃 𝑑
𝐺 𝜃 𝑔
𝑧
Objective Function
Noise
z
Generative adversarial nets, Goodfellow, J Pouget-Abadie, M Mirza, B Xu, D Warde-Farley, S Ozair, NIPS 2014
CONDITIONAL GAN
Conditional GANs learn a mapping from observed image x and random noise vector z,
to y, G : {x, z} → y.
23Image-to-Image Translation with Conditional Adversarial Networks, P. Isola et al., CVPR 2017
GAN FOR DATA AUGMENTATION
Removing the need for dataset
collection with GAN-based
image-to-image transformation.
24
Reuse the detection
bounding boxes
25
PRESERVING IMAGE-OBJECTS FOR IMAGE TRANSLATION
Cycle GAN AugGAN
AugGAN: Cross Domain Adaptation with GAN-based Data Augmentation, S. Huang et al., ECCV 2018
ACCURACY COMPARISON FOR OBJECT DETECTION
Comparison of detection accuracies for YOLO and Faster RCNN detectors trained with
transformed data by applying dierent GANs trained from SYNTHIA and GTA datasets.
KITTI-D2N-S/G: KITTI Day-to-Night training data transformed by GAN learnt from
SYNTHIA/GTA
SCNT: Self-Collected Nighttime Testing data.
26AugGAN: Cross Domain Adaptation with GAN-based Data Augmentation, S. Huang et al., ECCV 2018
SYNTHESIZING DATA FOR TRAINING DNN
27
WHY TRAINING
DNN WITH
SYNTHETIC DATA?
28
It’s easy to produce data once a synthetic
model/environment has been established.
Accurate label is expensive, sometimes difficult to
get, or labor intensive
Synthetic environment is flexible to be adjusted as
needed.
Synthetic data is a substitute for data that contains
sensitive information
Synthetic data has been popularly used in areas
like medical imaging, autonomous driving, robotic
control/navigation
WAFER MAP DEFECT PATTERN CLASSIFICATION
Wafer Map Defect Pattern Classification and Image Retrieval Using Convolutional Neural Network, IEEE Transactions on
Semiconductor Manufacturing, 2018
• Generate synthetic wafer maps through Poisson point process.
• Testing on real wafer maps
Classification accuracy: almost 100%
Image Retrieval Error Rate: 3.7%
Examples of the generated wafer
maps of different classes
29
WRENCH DETECTION
Generate synthetic images in virtual environment.
• Randomly adjust the position and texture of each wrench.
• Automatically generate accurate segmentation mask.
Realdata
3D wrench
model
Segmentation
mask
Complete-wrenches-
only mask
Synthetic
environment
Image generated
by Unity
30
WRENCH DETECTION AND SEGMENTATION
Photorealism
• Randomly add gaussian noise to generated image.
• Transfer the style of generated images into the style of real data.
Style
Transfer*
Real Data
Images shot in
Unity
Final synthetic
image
Gaussian Noise
*A Closed-form Solution to Photorealistic Image Stylization, ECCV 2018 31
SYNTHETIC DATA
Before After Before After Before After
WRENCH DETECTION
Average Precision @0.5 IoU of
using different training data for the wrench detection task.
Real data: 392 images, Synthetic data: 1,000 images.
Training Data Mask AP Bbox AP
Only Real 57.1 72.0
Only Synthetic (w/o style transfer) 54.4 60.4
Only Synthetic (w/ style transfer) 59.0 64.7
Real and Synthetic (w/o style transfer) 73.0 76.9
Real and Synthetic (w/ style transfer) 78.2 82.0
Using Mask R-CNN as the wrench detector.
Y.-H. Lee et al., Automatic generation of photorealistic training data for detection of industrial components”, ICIP 2019
COMPUTER VISION SERVICES
Office
 SharePoint + OneDrive
 PPTX + Word Accessibility
 Detail
Cognitive Services
 Video Indexer
 Content Moderator
Custom Vision
 Image Classification
 Object Detection
OCR
 Leading OCR model
 Forms + Structured documents
 Exposed via Computer Vision API
Computer Vision
 Image Tagging
 Image Captioning
 Adult, Logos, etc
Face
 Face Recognition
 Face Detection
 Attributes
PartnersInvestments
Dynamics
• Retail
• Mixed Reality
Form Recognizer
• Extract text, key-value pairs, and tables
• Customized to your forms, without manual
labeling
CUSTOM VISION SERVICE
CUSTOM VISION SERVICE
https://www.customvision.ai
TRAIN IN THE CLOUD, RUN ANYWHERE
Client Platform Format
iOS CoreML
Android TensorFlow, TensorFlow Lite
Docker,
Azure IOT Edge, Azure
Functions, Azure ML
Linux, Windows, ARM
Windows ONNX
DOCKER CONTAINER EXPORT
You own
Prediction Endpoint
(Docker container)
Local or on-prem
Azure App Service
CUSTOM VISION SERVICE
• Current:
• Robust object detectors and image classifiers with
fast training speed and advanced training option
• Train on cloud, evaluate on device:
https://customvision.ai -> Android, iOS, ONNX,
Docker or Cloud service
• Real Customer Scenarios:
• Visual alerts from IoT cameras (ie workplace safety,
truck load detection, traffic)
• Product recognition for grocery store check-out
• Object counting
• Social media analysis (logo detection)
• Drone imagery analysis
Product counting
Logo recognition
Pedestrian, cars detection
SAMPLE SOLUTION
https://github.com/Azure-Samples/Cognitive-
Services-Vision-Solution-Templates
Custom Vision Service and Azure IOT Hub
TRY IT YOURSELF
CONCLUSION
➢The performance of the model is heavily dependent on the training data.
➢Sample quantity, annotation quality, and representative variation of the
training data are all very critical to the success of an AI system.
➢Data augmentation is a simple way to increase the data size.
➢Data annotation/labeling efforts can be dramatically reduced semi-
supervised learning.
➢Data synthesis has been proved to be quite successful for training DNN
models in many real-world applications.
➢Domain adaptation is very useful to make sure the distribution of training
data similar to that of real application scenario.
42

More Related Content

What's hot

2 клас 23 урок. Графічний редактор Tux Paint.
2 клас 23 урок. Графічний редактор Tux Paint.2 клас 23 урок. Графічний редактор Tux Paint.
2 клас 23 урок. Графічний редактор Tux Paint.StAlKeRoV
 
99個簡報法則
99個簡報法則99個簡報法則
99個簡報法則奕酉 劉
 
Професійна діяльність і особистість педагога
Професійна діяльність і особистість педагогаПрофесійна діяльність і особистість педагога
Професійна діяльність і особистість педагогаTonya777
 
TaxiGo LINE Chatbot 開發歷程
TaxiGo LINE Chatbot 開發歷程TaxiGo LINE Chatbot 開發歷程
TaxiGo LINE Chatbot 開發歷程LINE Corporation
 
[系列活動] 智慧製造與生產線上的資料科學 (製造資料科學:從預測性思維到處方性決策)
[系列活動] 智慧製造與生產線上的資料科學 (製造資料科學:從預測性思維到處方性決策)[系列活動] 智慧製造與生產線上的資料科學 (製造資料科學:從預測性思維到處方性決策)
[系列活動] 智慧製造與生產線上的資料科學 (製造資料科學:從預測性思維到處方性決策)台灣資料科學年會
 
Методика навчання мистецтва у початковій школі Посібник для вчителів JI. М. ...
Методика навчання мистецтва у початковій школі Посібник для вчителів  JI. М. ...Методика навчання мистецтва у початковій школі Посібник для вчителів  JI. М. ...
Методика навчання мистецтва у початковій школі Посібник для вчителів JI. М. ...Ковпитська ЗОШ
 
商業模式創新講座簡報 110624
商業模式創新講座簡報 110624商業模式創新講座簡報 110624
商業模式創新講座簡報 110624基欽 劉
 
高度客製化Aoi系統的未來趨勢與挑戰
高度客製化Aoi系統的未來趨勢與挑戰高度客製化Aoi系統的未來趨勢與挑戰
高度客製化Aoi系統的未來趨勢與挑戰COMOSS Electronic Co., Ltd.
 
數位轉型的困難與挑戰
數位轉型的困難與挑戰數位轉型的困難與挑戰
數位轉型的困難與挑戰Charles Lee
 
2023-04-06 AI晶片產業整合版 研究部20230406.pdf
2023-04-06 AI晶片產業整合版 研究部20230406.pdf2023-04-06 AI晶片產業整合版 研究部20230406.pdf
2023-04-06 AI晶片產業整合版 研究部20230406.pdfJoyceHsu27
 
2 використання ікт в процесі самоосвіти педагога малахов а.а.
2 використання ікт в процесі самоосвіти педагога малахов а.а.2 використання ікт в процесі самоосвіти педагога малахов а.а.
2 використання ікт в процесі самоосвіти педагога малахов а.а.Андррей Малахов
 
конструювання компетентніснних завдань
конструювання компетентніснних завданьконструювання компетентніснних завдань
конструювання компетентніснних завданьРоман Бадулін
 
1.2 Двійкове кодування. Одиниці вимірювання довжини двійкового коду. Кодуванн...
1.2 Двійкове кодування. Одиниці вимірювання довжини двійкового коду. Кодуванн...1.2 Двійкове кодування. Одиниці вимірювання довжини двійкового коду. Кодуванн...
1.2 Двійкове кодування. Одиниці вимірювання довжини двійкового коду. Кодуванн...Оксана Миколаївна Хоровець
 
Педагогічна рада на тему «Формування здоров’язберігаючих компетентностей шко...
Педагогічна рада на тему «Формування здоров’язберігаючих компетентностей  шко...Педагогічна рада на тему «Формування здоров’язберігаючих компетентностей  шко...
Педагогічна рада на тему «Формування здоров’язберігаючих компетентностей шко...Dmytro Kryvenko
 
Календарне планування
Календарне плануванняКалендарне планування
Календарне плануванняIrina Shlyahetskaya
 
Школа молодого вчителя
Школа молодого вчителяШкола молодого вчителя
Школа молодого вчителяJoey Badass
 
Математика. 2 клас. Методика підготовчої роботи і ознайомлення учнів 2-го кла...
Математика. 2 клас. Методика підготовчої роботи і ознайомлення учнів 2-го кла...Математика. 2 клас. Методика підготовчої роботи і ознайомлення учнів 2-го кла...
Математика. 2 клас. Методика підготовчої роботи і ознайомлення учнів 2-го кла...Електронні книги Ранок
 
«Причини невисокої якості знань учнів і шляхи їх подолання»
«Причини  невисокої якості знань учнів і шляхи їх подолання» «Причини  невисокої якості знань учнів і шляхи їх подолання»
«Причини невисокої якості знань учнів і шляхи їх подолання» sasha884
 

What's hot (20)

Початкова школа: освіта для життя
Початкова школа: освіта для життяПочаткова школа: освіта для життя
Початкова школа: освіта для життя
 
2 клас 23 урок. Графічний редактор Tux Paint.
2 клас 23 урок. Графічний редактор Tux Paint.2 клас 23 урок. Графічний редактор Tux Paint.
2 клас 23 урок. Графічний редактор Tux Paint.
 
99個簡報法則
99個簡報法則99個簡報法則
99個簡報法則
 
Професійна діяльність і особистість педагога
Професійна діяльність і особистість педагогаПрофесійна діяльність і особистість педагога
Професійна діяльність і особистість педагога
 
TaxiGo LINE Chatbot 開發歷程
TaxiGo LINE Chatbot 開發歷程TaxiGo LINE Chatbot 開發歷程
TaxiGo LINE Chatbot 開發歷程
 
[系列活動] 智慧製造與生產線上的資料科學 (製造資料科學:從預測性思維到處方性決策)
[系列活動] 智慧製造與生產線上的資料科學 (製造資料科學:從預測性思維到處方性決策)[系列活動] 智慧製造與生產線上的資料科學 (製造資料科學:從預測性思維到處方性決策)
[系列活動] 智慧製造與生產線上的資料科學 (製造資料科學:從預測性思維到處方性決策)
 
Методика навчання мистецтва у початковій школі Посібник для вчителів JI. М. ...
Методика навчання мистецтва у початковій школі Посібник для вчителів  JI. М. ...Методика навчання мистецтва у початковій школі Посібник для вчителів  JI. М. ...
Методика навчання мистецтва у початковій школі Посібник для вчителів JI. М. ...
 
商業模式創新講座簡報 110624
商業模式創新講座簡報 110624商業模式創新講座簡報 110624
商業模式創新講座簡報 110624
 
Педрада Бессараб С.О.
Педрада Бессараб С.О.Педрада Бессараб С.О.
Педрада Бессараб С.О.
 
高度客製化Aoi系統的未來趨勢與挑戰
高度客製化Aoi系統的未來趨勢與挑戰高度客製化Aoi系統的未來趨勢與挑戰
高度客製化Aoi系統的未來趨勢與挑戰
 
數位轉型的困難與挑戰
數位轉型的困難與挑戰數位轉型的困難與挑戰
數位轉型的困難與挑戰
 
2023-04-06 AI晶片產業整合版 研究部20230406.pdf
2023-04-06 AI晶片產業整合版 研究部20230406.pdf2023-04-06 AI晶片產業整合版 研究部20230406.pdf
2023-04-06 AI晶片產業整合版 研究部20230406.pdf
 
2 використання ікт в процесі самоосвіти педагога малахов а.а.
2 використання ікт в процесі самоосвіти педагога малахов а.а.2 використання ікт в процесі самоосвіти педагога малахов а.а.
2 використання ікт в процесі самоосвіти педагога малахов а.а.
 
конструювання компетентніснних завдань
конструювання компетентніснних завданьконструювання компетентніснних завдань
конструювання компетентніснних завдань
 
1.2 Двійкове кодування. Одиниці вимірювання довжини двійкового коду. Кодуванн...
1.2 Двійкове кодування. Одиниці вимірювання довжини двійкового коду. Кодуванн...1.2 Двійкове кодування. Одиниці вимірювання довжини двійкового коду. Кодуванн...
1.2 Двійкове кодування. Одиниці вимірювання довжини двійкового коду. Кодуванн...
 
Педагогічна рада на тему «Формування здоров’язберігаючих компетентностей шко...
Педагогічна рада на тему «Формування здоров’язберігаючих компетентностей  шко...Педагогічна рада на тему «Формування здоров’язберігаючих компетентностей  шко...
Педагогічна рада на тему «Формування здоров’язберігаючих компетентностей шко...
 
Календарне планування
Календарне плануванняКалендарне планування
Календарне планування
 
Школа молодого вчителя
Школа молодого вчителяШкола молодого вчителя
Школа молодого вчителя
 
Математика. 2 клас. Методика підготовчої роботи і ознайомлення учнів 2-го кла...
Математика. 2 клас. Методика підготовчої роботи і ознайомлення учнів 2-го кла...Математика. 2 клас. Методика підготовчої роботи і ознайомлення учнів 2-го кла...
Математика. 2 клас. Методика підготовчої роботи і ознайомлення учнів 2-го кла...
 
«Причини невисокої якості знань учнів і шляхи їх подолання»
«Причини  невисокої якості знань учнів і шляхи їх подолання» «Причини  невисокої якості знань учнів і шляхи їх подолання»
«Причини невисокої якості знань учнів і шляхи їх подолання»
 

Similar to 小數據如何實現電腦視覺

A Literature Survey on Image Linguistic Visual Question Answering
A Literature Survey on Image Linguistic Visual Question AnsweringA Literature Survey on Image Linguistic Visual Question Answering
A Literature Survey on Image Linguistic Visual Question AnsweringIRJET Journal
 
AI in the Financial Services Industry
AI in the Financial Services IndustryAI in the Financial Services Industry
AI in the Financial Services IndustryAlison B. Lowndes
 
Real-Time Pertinent Maneuver Recognition for Surveillance
Real-Time Pertinent Maneuver Recognition for SurveillanceReal-Time Pertinent Maneuver Recognition for Surveillance
Real-Time Pertinent Maneuver Recognition for SurveillanceIRJET Journal
 
Machine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdfMachine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdfCarlos Paredes
 
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde..."Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...Edge AI and Vision Alliance
 
Obscenity Detection in Images
Obscenity Detection in ImagesObscenity Detection in Images
Obscenity Detection in ImagesAnil Kumar Gupta
 
IRJET- Object Detection and Recognition using Single Shot Multi-Box Detector
IRJET- Object Detection and Recognition using Single Shot Multi-Box DetectorIRJET- Object Detection and Recognition using Single Shot Multi-Box Detector
IRJET- Object Detection and Recognition using Single Shot Multi-Box DetectorIRJET Journal
 
CAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNINGCAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNINGIRJET Journal
 
Object Detetcion using SSD-MobileNet
Object Detetcion using SSD-MobileNetObject Detetcion using SSD-MobileNet
Object Detetcion using SSD-MobileNetIRJET Journal
 
深度學習在AOI的應用
深度學習在AOI的應用深度學習在AOI的應用
深度學習在AOI的應用CHENHuiMei
 
IRJET- Wearable AI Device for Blind
IRJET- Wearable AI Device for BlindIRJET- Wearable AI Device for Blind
IRJET- Wearable AI Device for BlindIRJET Journal
 
IRJET - Autonomous Navigation System using Deep Learning
IRJET -  	  Autonomous Navigation System using Deep LearningIRJET -  	  Autonomous Navigation System using Deep Learning
IRJET - Autonomous Navigation System using Deep LearningIRJET Journal
 
B4UConference_machine learning_deeplearning
B4UConference_machine learning_deeplearningB4UConference_machine learning_deeplearning
B4UConference_machine learning_deeplearningHoa Le
 
Dataset creation for Deep Learning-based Geometric Computer Vision problems
Dataset creation for Deep Learning-based Geometric Computer Vision problemsDataset creation for Deep Learning-based Geometric Computer Vision problems
Dataset creation for Deep Learning-based Geometric Computer Vision problemsPetteriTeikariPhD
 
Creating a Machine Learning Model on the Cloud
Creating a Machine Learning Model on the CloudCreating a Machine Learning Model on the Cloud
Creating a Machine Learning Model on the CloudAlexander Al Basosi
 
IRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia
IRJET- A Survey on Medical Image Interpretation for Predicting PneumoniaIRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia
IRJET- A Survey on Medical Image Interpretation for Predicting PneumoniaIRJET Journal
 
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr TeterwakLearn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr TeterwakPyData
 

Similar to 小數據如何實現電腦視覺 (20)

A Literature Survey on Image Linguistic Visual Question Answering
A Literature Survey on Image Linguistic Visual Question AnsweringA Literature Survey on Image Linguistic Visual Question Answering
A Literature Survey on Image Linguistic Visual Question Answering
 
AI in the Financial Services Industry
AI in the Financial Services IndustryAI in the Financial Services Industry
AI in the Financial Services Industry
 
Real-Time Pertinent Maneuver Recognition for Surveillance
Real-Time Pertinent Maneuver Recognition for SurveillanceReal-Time Pertinent Maneuver Recognition for Surveillance
Real-Time Pertinent Maneuver Recognition for Surveillance
 
Machine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdfMachine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdf
 
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde..."Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
 
AI and Deep Learning
AI and Deep Learning AI and Deep Learning
AI and Deep Learning
 
Obscenity Detection in Images
Obscenity Detection in ImagesObscenity Detection in Images
Obscenity Detection in Images
 
IRJET- Object Detection and Recognition using Single Shot Multi-Box Detector
IRJET- Object Detection and Recognition using Single Shot Multi-Box DetectorIRJET- Object Detection and Recognition using Single Shot Multi-Box Detector
IRJET- Object Detection and Recognition using Single Shot Multi-Box Detector
 
kanimozhi2019.pdf
kanimozhi2019.pdfkanimozhi2019.pdf
kanimozhi2019.pdf
 
OpenPOWER/POWER9 AI webinar
OpenPOWER/POWER9 AI webinar OpenPOWER/POWER9 AI webinar
OpenPOWER/POWER9 AI webinar
 
CAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNINGCAR DAMAGE DETECTION USING DEEP LEARNING
CAR DAMAGE DETECTION USING DEEP LEARNING
 
Object Detetcion using SSD-MobileNet
Object Detetcion using SSD-MobileNetObject Detetcion using SSD-MobileNet
Object Detetcion using SSD-MobileNet
 
深度學習在AOI的應用
深度學習在AOI的應用深度學習在AOI的應用
深度學習在AOI的應用
 
IRJET- Wearable AI Device for Blind
IRJET- Wearable AI Device for BlindIRJET- Wearable AI Device for Blind
IRJET- Wearable AI Device for Blind
 
IRJET - Autonomous Navigation System using Deep Learning
IRJET -  	  Autonomous Navigation System using Deep LearningIRJET -  	  Autonomous Navigation System using Deep Learning
IRJET - Autonomous Navigation System using Deep Learning
 
B4UConference_machine learning_deeplearning
B4UConference_machine learning_deeplearningB4UConference_machine learning_deeplearning
B4UConference_machine learning_deeplearning
 
Dataset creation for Deep Learning-based Geometric Computer Vision problems
Dataset creation for Deep Learning-based Geometric Computer Vision problemsDataset creation for Deep Learning-based Geometric Computer Vision problems
Dataset creation for Deep Learning-based Geometric Computer Vision problems
 
Creating a Machine Learning Model on the Cloud
Creating a Machine Learning Model on the CloudCreating a Machine Learning Model on the Cloud
Creating a Machine Learning Model on the Cloud
 
IRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia
IRJET- A Survey on Medical Image Interpretation for Predicting PneumoniaIRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia
IRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia
 
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr TeterwakLearn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
Learn to Build an App to Find Similar Images using Deep Learning- Piotr Teterwak
 

More from CHENHuiMei

產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉CHENHuiMei
 
基於少樣本深度學習之橡膠墊片檢測系統
基於少樣本深度學習之橡膠墊片檢測系統基於少樣本深度學習之橡膠墊片檢測系統
基於少樣本深度學習之橡膠墊片檢測系統CHENHuiMei
 
使用人工智慧檢測三維錫球瑕疵_台大傅楸善
使用人工智慧檢測三維錫球瑕疵_台大傅楸善使用人工智慧檢測三維錫球瑕疵_台大傅楸善
使用人工智慧檢測三維錫球瑕疵_台大傅楸善CHENHuiMei
 
IIoT發展趨勢及設備業者因應之_微軟葉怡君
IIoT發展趨勢及設備業者因應之_微軟葉怡君IIoT發展趨勢及設備業者因應之_微軟葉怡君
IIoT發展趨勢及設備業者因應之_微軟葉怡君CHENHuiMei
 
精密機械的空間軌跡精度光學檢測法_台大范光照
精密機械的空間軌跡精度光學檢測法_台大范光照精密機械的空間軌跡精度光學檢測法_台大范光照
精密機械的空間軌跡精度光學檢測法_台大范光照CHENHuiMei
 
When AOI meets AI
When AOI meets AIWhen AOI meets AI
When AOI meets AICHENHuiMei
 
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠CHENHuiMei
 
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士CHENHuiMei
 
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民CHENHuiMei
 
2018AOI論壇_AOI and IoT產線應用_工研院周森益
2018AOI論壇_AOI and IoT產線應用_工研院周森益2018AOI論壇_AOI and IoT產線應用_工研院周森益
2018AOI論壇_AOI and IoT產線應用_工研院周森益CHENHuiMei
 
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢CHENHuiMei
 
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章CHENHuiMei
 
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏CHENHuiMei
 
200704 Dr. Schenk 產品策略
200704 Dr. Schenk 產品策略200704 Dr. Schenk 產品策略
200704 Dr. Schenk 產品策略CHENHuiMei
 
2007 TFT LCD-AOI教學檔案
2007 TFT LCD-AOI教學檔案2007 TFT LCD-AOI教學檔案
2007 TFT LCD-AOI教學檔案CHENHuiMei
 
2008 year Computer Vision
2008 year Computer Vision2008 year Computer Vision
2008 year Computer VisionCHENHuiMei
 
2008 tech 高分子微光學應用技術
2008 tech 高分子微光學應用技術2008 tech 高分子微光學應用技術
2008 tech 高分子微光學應用技術CHENHuiMei
 
2008 tech Instrumentation in Precision Metrology
2008 tech Instrumentation in Precision Metrology2008 tech Instrumentation in Precision Metrology
2008 tech Instrumentation in Precision MetrologyCHENHuiMei
 

More from CHENHuiMei (20)

產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
 
基於少樣本深度學習之橡膠墊片檢測系統
基於少樣本深度學習之橡膠墊片檢測系統基於少樣本深度學習之橡膠墊片檢測系統
基於少樣本深度學習之橡膠墊片檢測系統
 
使用人工智慧檢測三維錫球瑕疵_台大傅楸善
使用人工智慧檢測三維錫球瑕疵_台大傅楸善使用人工智慧檢測三維錫球瑕疵_台大傅楸善
使用人工智慧檢測三維錫球瑕疵_台大傅楸善
 
IIoT發展趨勢及設備業者因應之_微軟葉怡君
IIoT發展趨勢及設備業者因應之_微軟葉怡君IIoT發展趨勢及設備業者因應之_微軟葉怡君
IIoT發展趨勢及設備業者因應之_微軟葉怡君
 
精密機械的空間軌跡精度光學檢測法_台大范光照
精密機械的空間軌跡精度光學檢測法_台大范光照精密機械的空間軌跡精度光學檢測法_台大范光照
精密機械的空間軌跡精度光學檢測法_台大范光照
 
Report
ReportReport
Report
 
Deep learning
Deep learningDeep learning
Deep learning
 
When AOI meets AI
When AOI meets AIWhen AOI meets AI
When AOI meets AI
 
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
 
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
 
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
 
2018AOI論壇_AOI and IoT產線應用_工研院周森益
2018AOI論壇_AOI and IoT產線應用_工研院周森益2018AOI論壇_AOI and IoT產線應用_工研院周森益
2018AOI論壇_AOI and IoT產線應用_工研院周森益
 
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
 
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
 
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
 
200704 Dr. Schenk 產品策略
200704 Dr. Schenk 產品策略200704 Dr. Schenk 產品策略
200704 Dr. Schenk 產品策略
 
2007 TFT LCD-AOI教學檔案
2007 TFT LCD-AOI教學檔案2007 TFT LCD-AOI教學檔案
2007 TFT LCD-AOI教學檔案
 
2008 year Computer Vision
2008 year Computer Vision2008 year Computer Vision
2008 year Computer Vision
 
2008 tech 高分子微光學應用技術
2008 tech 高分子微光學應用技術2008 tech 高分子微光學應用技術
2008 tech 高分子微光學應用技術
 
2008 tech Instrumentation in Precision Metrology
2008 tech Instrumentation in Precision Metrology2008 tech Instrumentation in Precision Metrology
2008 tech Instrumentation in Precision Metrology
 

Recently uploaded

Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Mark Simos
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxNavinnSomaal
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Commit University
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyAlfredo García Lavilla
 
Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piececharlottematthew16
 
My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024The Digital Insurer
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitecturePixlogix Infotech
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Wonjun Hwang
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):comworks
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii SoldatenkoFwdays
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...Fwdays
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Enterprise Knowledge
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxhariprasad279825
 
APIForce Zurich 5 April Automation LPDG
APIForce Zurich 5 April  Automation LPDGAPIForce Zurich 5 April  Automation LPDG
APIForce Zurich 5 April Automation LPDGMarianaLemus7
 

Recently uploaded (20)

Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptx
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easy
 
Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piece
 
Hot Sexy call girls in Panjabi Bagh 🔝 9953056974 🔝 Delhi escort Service
Hot Sexy call girls in Panjabi Bagh 🔝 9953056974 🔝 Delhi escort ServiceHot Sexy call girls in Panjabi Bagh 🔝 9953056974 🔝 Delhi escort Service
Hot Sexy call girls in Panjabi Bagh 🔝 9953056974 🔝 Delhi escort Service
 
My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):
 
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptxE-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptx
 
APIForce Zurich 5 April Automation LPDG
APIForce Zurich 5 April  Automation LPDGAPIForce Zurich 5 April  Automation LPDG
APIForce Zurich 5 April Automation LPDG
 

小數據如何實現電腦視覺

  • 1. 小數據如何實現電腦視覺 賴尚宏 1 Principal Researcher Microsoft AI R&D Center Taipei Professor Dept. of Computer Science National Tsing Hua University
  • 2. DEEP LEARNING → COMPUTER VISION ➢Success in applying deep learning to a number of computer vision problems ➢Image classification ➢Image segmentation ➢Object detection, tracking, recognition ➢Image processing (style transfer, super-resolution, deblurring, etc.) ➢3D reconstruction ➢Image/video captioning/Q&A ➢……. 2
  • 3. CNN (CONVOLUTIONAL NEURAL NETWORKS) CNN has been successfully used in many image classification and object detection problems. The most famous CNN architecture is LeNet used to classify hand-written digits. Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner. Gradient-based learning applied to document recognition. Proc. of IEEE, 1998
  • 4. MODEL COMPLEXITY The complexity of a learning model can be characterized by the number of parameters in the model. A large number of model parameters means it can learn/model very complex structure. Recent deep networks usually contain millions of parameters to be learned from the training data. 4Figures adopted from Deep Learning, by I. Goodfellow, Y. Bengio and A. Courville, MIT Press
  • 5. DATASET SIZE 5 The age of “Big Data” has made machine learning much easier because it provides a very large number of training data of wide varieties for the deep networks (millions of parameters) to learn to achieve the level of human capability. Figures adopted from Deep Learning, by I. Goodfellow, Y. Bengio and A. Courville, MIT Press
  • 6. FACE RECOGNITION DATASET Dataset Subjects Images Note CASIA-WebFace 10,575 494,414 The second largest public dataset available for face verification and recognition problems. VGGFace2 9131 3.31 million Images have large variations in pose, age, illumination, ethnicity and profession. MS-Celeb-1M 100k 10 million To facilitate the IRC face recognition task LFW 5749 13,233 For face verification FaceScrub 530 107,818 One of the largest public face databases. MegaFace 672,000+ 4.7 million the largest (in number of identities) publicly available facial recognition dataset CelebA 10,177 202,599 5 landmark locations, 40 binary attributes annotations per image. MultiPie 337 750,000+ Under 15 view points and 19 illumination conditions UMDFaces 8,277 367,888 Contains both still images and video frames.
  • 7. DATA ANNOTATION ➢Image-level annotation ➢Object-level annotation 7 { edge, center } { edge (point), center (point) } { edge (b-box), center (b-box) } { edge (pixel labels), center (pixel labels) } very fast fast slow very slow
  • 8. PROBLEMS WITH PREPARING TRAINING DATA ➢Require large amount of training data of wide varieties ➢Imbalanced data ➢Usually domain specific ➢Intensive human labeling cost ➢Restricted to supervised learning 8
  • 10. DATA AUGMENTATION State-of-the-art neural networks typically have parameters in the order of millions. The number of parameters needed is proportional to the complexity of the task your model has to perform. The size of data samples for training the model should be proportional to the number of parameters of the model. 10
  • 12. Data Augmentation for Training Detector of Partially Occluded Cars Augmentation strategy: resize the vehicles in the training images with 0.1 to 0.9 of the original size horizontally flip the images with 0.5 probability. truncate a vehicle with 0.5 probability by a new window such that the vehicle contains at least 25% of its appearance visible. 12
  • 14. TRANSFER LEARNING 14 Transfer learning is a machine learning method where a model trained for a task is reused as the starting point for re-training a model for another task. It’s popular for some computer vision tasks that require vast compute and time resources to train large and deep neural network models. This learning process will tend to work if the features are general, meaning suitable to both base and target tasks, instead of specific to the base task.
  • 15. TRANSFER LEARNING WITH CNNS 15Slide credit: Stanford Univ. CS 231n
  • 16. GENERAL GUIDELINE FOR FINETUNING DEEP NEURAL NETWORKS 16Slide credit: Stanford Univ. CS 231n
  • 18. INTRODUCTION Supervised Learning – Train a model with labeled data only. Semi-Supervised Learning – Train a model with labeled(R) and unlabeled(U) data, usually U>>R. Why semi-supervised learning ? Collecting data is easy, but collecting labelled data is expensive. 18 Some Labeled Data Lots of Unlabeled Data Model Semi-supervised Learning
  • 19. PSEUDO LABELING 19 Example: Cat or Dog Classification DOG CAT Labeled Data Unlabeled Data 1/5 DOG CAT Labeled Data 2/5 TRAIN Unlabeled Data 3/5 PREDICTPREDICT DOG CAT 4/5 TRAIN 5/5
  • 20. DISCUSSION Collecting labelled data is expensive More accurate decision boundary with unlabeled instances 20 Decision boundary Labeled Instances Decision boundary Unlabeled Instances • No way to verify the produced labels’ accuracy → less trust worthy Pros Cons
  • 21. GAN FOR IMAGE SYNTHESIS (Generative Adversarial Network) 21
  • 22. GAN ARCHITECTURE 22 Real or Fake Loss G Generator Network D Discriminator Network Fake Images G(z) Real Images x min 𝜃 𝑔 max 𝜃 𝑑 ቈ ቉ 𝔼 𝑥∼𝑝 𝑑𝑎𝑡𝑎 log 𝐷 𝜃 𝑑 𝑥 + 𝔼 𝑧∼𝑝 𝑧 log ቆ ቇ 1 − 𝐷 𝜃 𝑑 𝐺 𝜃 𝑔 𝑧 Objective Function Noise z Generative adversarial nets, Goodfellow, J Pouget-Abadie, M Mirza, B Xu, D Warde-Farley, S Ozair, NIPS 2014
  • 23. CONDITIONAL GAN Conditional GANs learn a mapping from observed image x and random noise vector z, to y, G : {x, z} → y. 23Image-to-Image Translation with Conditional Adversarial Networks, P. Isola et al., CVPR 2017
  • 24. GAN FOR DATA AUGMENTATION Removing the need for dataset collection with GAN-based image-to-image transformation. 24 Reuse the detection bounding boxes
  • 25. 25 PRESERVING IMAGE-OBJECTS FOR IMAGE TRANSLATION Cycle GAN AugGAN AugGAN: Cross Domain Adaptation with GAN-based Data Augmentation, S. Huang et al., ECCV 2018
  • 26. ACCURACY COMPARISON FOR OBJECT DETECTION Comparison of detection accuracies for YOLO and Faster RCNN detectors trained with transformed data by applying dierent GANs trained from SYNTHIA and GTA datasets. KITTI-D2N-S/G: KITTI Day-to-Night training data transformed by GAN learnt from SYNTHIA/GTA SCNT: Self-Collected Nighttime Testing data. 26AugGAN: Cross Domain Adaptation with GAN-based Data Augmentation, S. Huang et al., ECCV 2018
  • 27. SYNTHESIZING DATA FOR TRAINING DNN 27
  • 28. WHY TRAINING DNN WITH SYNTHETIC DATA? 28 It’s easy to produce data once a synthetic model/environment has been established. Accurate label is expensive, sometimes difficult to get, or labor intensive Synthetic environment is flexible to be adjusted as needed. Synthetic data is a substitute for data that contains sensitive information Synthetic data has been popularly used in areas like medical imaging, autonomous driving, robotic control/navigation
  • 29. WAFER MAP DEFECT PATTERN CLASSIFICATION Wafer Map Defect Pattern Classification and Image Retrieval Using Convolutional Neural Network, IEEE Transactions on Semiconductor Manufacturing, 2018 • Generate synthetic wafer maps through Poisson point process. • Testing on real wafer maps Classification accuracy: almost 100% Image Retrieval Error Rate: 3.7% Examples of the generated wafer maps of different classes 29
  • 30. WRENCH DETECTION Generate synthetic images in virtual environment. • Randomly adjust the position and texture of each wrench. • Automatically generate accurate segmentation mask. Realdata 3D wrench model Segmentation mask Complete-wrenches- only mask Synthetic environment Image generated by Unity 30
  • 31. WRENCH DETECTION AND SEGMENTATION Photorealism • Randomly add gaussian noise to generated image. • Transfer the style of generated images into the style of real data. Style Transfer* Real Data Images shot in Unity Final synthetic image Gaussian Noise *A Closed-form Solution to Photorealistic Image Stylization, ECCV 2018 31
  • 32. SYNTHETIC DATA Before After Before After Before After
  • 33. WRENCH DETECTION Average Precision @0.5 IoU of using different training data for the wrench detection task. Real data: 392 images, Synthetic data: 1,000 images. Training Data Mask AP Bbox AP Only Real 57.1 72.0 Only Synthetic (w/o style transfer) 54.4 60.4 Only Synthetic (w/ style transfer) 59.0 64.7 Real and Synthetic (w/o style transfer) 73.0 76.9 Real and Synthetic (w/ style transfer) 78.2 82.0 Using Mask R-CNN as the wrench detector. Y.-H. Lee et al., Automatic generation of photorealistic training data for detection of industrial components”, ICIP 2019
  • 34. COMPUTER VISION SERVICES Office  SharePoint + OneDrive  PPTX + Word Accessibility  Detail Cognitive Services  Video Indexer  Content Moderator Custom Vision  Image Classification  Object Detection OCR  Leading OCR model  Forms + Structured documents  Exposed via Computer Vision API Computer Vision  Image Tagging  Image Captioning  Adult, Logos, etc Face  Face Recognition  Face Detection  Attributes PartnersInvestments Dynamics • Retail • Mixed Reality Form Recognizer • Extract text, key-value pairs, and tables • Customized to your forms, without manual labeling
  • 37. TRAIN IN THE CLOUD, RUN ANYWHERE Client Platform Format iOS CoreML Android TensorFlow, TensorFlow Lite Docker, Azure IOT Edge, Azure Functions, Azure ML Linux, Windows, ARM Windows ONNX
  • 38. DOCKER CONTAINER EXPORT You own Prediction Endpoint (Docker container) Local or on-prem Azure App Service
  • 39. CUSTOM VISION SERVICE • Current: • Robust object detectors and image classifiers with fast training speed and advanced training option • Train on cloud, evaluate on device: https://customvision.ai -> Android, iOS, ONNX, Docker or Cloud service • Real Customer Scenarios: • Visual alerts from IoT cameras (ie workplace safety, truck load detection, traffic) • Product recognition for grocery store check-out • Object counting • Social media analysis (logo detection) • Drone imagery analysis Product counting Logo recognition Pedestrian, cars detection
  • 42. CONCLUSION ➢The performance of the model is heavily dependent on the training data. ➢Sample quantity, annotation quality, and representative variation of the training data are all very critical to the success of an AI system. ➢Data augmentation is a simple way to increase the data size. ➢Data annotation/labeling efforts can be dramatically reduced semi- supervised learning. ➢Data synthesis has been proved to be quite successful for training DNN models in many real-world applications. ➢Domain adaptation is very useful to make sure the distribution of training data similar to that of real application scenario. 42