SlideShare ist ein Scribd-Unternehmen logo
1 von 20
AdWorld :
Ad’s Based on Image
Chetan Kulkarni
San Jose State University
Deep Learning
Introduction
• With the world getting completely online, Advertisers must
innovate to increase the sales of product.
• The idea at hand was using the continuous data (Image) and
analyzing the image and producing ad’s
Dasetset Collection
1. Used Googles OpenImagesV6 Dataset
2. Leveraged openimages PyPi API to download dataset
required for the object detection.
3. Date Composition was image data and its specific
annotated data for that object
Data Example
Annotation in Pascal Format Image
Every image will have its annotation xml file , these both will be used as data for transfer learning
Data Example
Annotation in Pascal Format Image
Object Detection : Techniques
• In this project I have tried to use various Pretrained models to
perform Object Detection and on that Generating Ad’s.
• RetinaNet
• Yolov3
• YoloV3 tiny
• Transfer Learning on Yolov3
RetinaNet object Detection
Model built on COCO dataset, and supported following image detection:
RetinaNet object Detection
RetinaNet object Detection
Now Need to Generate Ad’s
Object
Object Category
Advertisement
The object will be detected by models
Admin will make Object Category,
Each object will be mapped to
one or more categories
Ad’s will be made to target
specific category
Django Application • TV is a object
• TV is connected to utilities
category
• Subwoofer Ad is targeted to
utilities category
• So whenever we have TV,
we show all the Ad’s of
utilities category
• Model Detected TV in the
image
• We have Advertisements of
Subwoofer and couch
connected to TV
• Hence The Ad’s
Transfer Learning
• Using Existing Yolov3 pretrained model. Tried to transfer learn for 3
specific imageset of table, clothing and bottle.
• Couldn’t Train model on basic google collab GPU’s
• Tried Google Collab PRO with High Ram and High Gpu
• It took 3000s for one epoch
• Because of limited resource couldn’t train for more than 24 hours
• Hit The Accuracy of 50% for these custom object detection
Transfer Learning
Transfer Learning: 3000s per epoch
Tensorboard
Tensorboard
Extending the detection model to Videos
• After the Image detection and Ad Generation on Images.
• Tried to Generate Ad’s on the Videos
• It Takes lot of processing and time.
• Splitting Videos into 20 Frames per image, and running image
detection model on them
• And on overall Image detected, Performing Ad Generation
Quick Video Output
Application is hosted on AWS
Conclusion
• Using Pretrained Models and Transfer learning we can take
image detection to next level
• Application of this image and object detection can be used in
different places like traffic
• Personalized Ads can be Generated based on specific user and
Ad’s can be served real time with live video feed in Mall’s

Weitere ähnliche Inhalte

Ähnlich wie Image-Based Ad Generation Using Deep Learning Object Detection

Real Time Object Dectection using machine learning
Real Time Object Dectection using machine learningReal Time Object Dectection using machine learning
Real Time Object Dectection using machine learningpratik pratyay
 
Ml goes fruitful
Ml goes fruitfulMl goes fruitful
Ml goes fruitfulPreeti Negi
 
Using Crowdsourced Images to Create Image Recognition Models with Analytics Z...
Using Crowdsourced Images to Create Image Recognition Models with Analytics Z...Using Crowdsourced Images to Create Image Recognition Models with Analytics Z...
Using Crowdsourced Images to Create Image Recognition Models with Analytics Z...Maurice Nsabimana
 
Developing an object detector solution with Azure Custom Vision .NET SDK
Developing an object detector solution with Azure Custom Vision .NET SDKDeveloping an object detector solution with Azure Custom Vision .NET SDK
Developing an object detector solution with Azure Custom Vision .NET SDKLuis Beltran
 
Final PPT.ppt about human detection and counting
Final PPT.ppt  about human detection and countingFinal PPT.ppt  about human detection and counting
Final PPT.ppt about human detection and countingArbazAhmad25
 
Elderly Assistance- Deep Learning Theme detection
Elderly Assistance- Deep Learning Theme detectionElderly Assistance- Deep Learning Theme detection
Elderly Assistance- Deep Learning Theme detectionTanvi Mittal
 
AI Food detector; A model of Generative adversarial network for food Classifier
AI Food detector; A model of Generative adversarial network for food ClassifierAI Food detector; A model of Generative adversarial network for food Classifier
AI Food detector; A model of Generative adversarial network for food Classifierjimmy majumder
 
Yoga pose detection using deep learning project PPT.pptx
Yoga pose detection using deep learning project PPT.pptxYoga pose detection using deep learning project PPT.pptx
Yoga pose detection using deep learning project PPT.pptxssuser4f92fb
 
"Semantic Indexing of Wearable Camera Images: Kids’Cam Concepts"
"Semantic Indexing of Wearable Camera Images: Kids’Cam Concepts""Semantic Indexing of Wearable Camera Images: Kids’Cam Concepts"
"Semantic Indexing of Wearable Camera Images: Kids’Cam Concepts"Alan Smeaton
 
DSC Cloud Study Jams
DSC Cloud Study JamsDSC Cloud Study Jams
DSC Cloud Study JamsKateGrupp
 
DreamPose: Fashion Image to Video Synthesis via Stable Diffusion
DreamPose: Fashion Image to Video Synthesis via Stable DiffusionDreamPose: Fashion Image to Video Synthesis via Stable Diffusion
DreamPose: Fashion Image to Video Synthesis via Stable Diffusiondrawais8
 
Object detection presentation
Object detection presentationObject detection presentation
Object detection presentationAshwinBicholiya
 
Automated_attendance_system_project.pptx
Automated_attendance_system_project.pptxAutomated_attendance_system_project.pptx
Automated_attendance_system_project.pptxNaveensai51
 
Watson Visual Recognition - NYC JavaSig
Watson Visual Recognition - NYC JavaSigWatson Visual Recognition - NYC JavaSig
Watson Visual Recognition - NYC JavaSigUpkar Lidder
 
Introduction talk to Computer Vision
Introduction talk to Computer Vision Introduction talk to Computer Vision
Introduction talk to Computer Vision Chen Sagiv
 
502021435-12345678Minor-Project-Ppt.pptx
502021435-12345678Minor-Project-Ppt.pptx502021435-12345678Minor-Project-Ppt.pptx
502021435-12345678Minor-Project-Ppt.pptxshrey4922
 
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine LearningMakine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine LearningAli Alkan
 
Easy path to machine learning (Spring 2020)
Easy path to machine learning (Spring 2020)Easy path to machine learning (Spring 2020)
Easy path to machine learning (Spring 2020)wesley chun
 

Ähnlich wie Image-Based Ad Generation Using Deep Learning Object Detection (20)

Real Time Object Dectection using machine learning
Real Time Object Dectection using machine learningReal Time Object Dectection using machine learning
Real Time Object Dectection using machine learning
 
Ml goes fruitful
Ml goes fruitfulMl goes fruitful
Ml goes fruitful
 
slide-171212080528.pptx
slide-171212080528.pptxslide-171212080528.pptx
slide-171212080528.pptx
 
Using Crowdsourced Images to Create Image Recognition Models with Analytics Z...
Using Crowdsourced Images to Create Image Recognition Models with Analytics Z...Using Crowdsourced Images to Create Image Recognition Models with Analytics Z...
Using Crowdsourced Images to Create Image Recognition Models with Analytics Z...
 
Developing an object detector solution with Azure Custom Vision .NET SDK
Developing an object detector solution with Azure Custom Vision .NET SDKDeveloping an object detector solution with Azure Custom Vision .NET SDK
Developing an object detector solution with Azure Custom Vision .NET SDK
 
Final PPT.ppt about human detection and counting
Final PPT.ppt  about human detection and countingFinal PPT.ppt  about human detection and counting
Final PPT.ppt about human detection and counting
 
Elderly Assistance- Deep Learning Theme detection
Elderly Assistance- Deep Learning Theme detectionElderly Assistance- Deep Learning Theme detection
Elderly Assistance- Deep Learning Theme detection
 
AI Food detector; A model of Generative adversarial network for food Classifier
AI Food detector; A model of Generative adversarial network for food ClassifierAI Food detector; A model of Generative adversarial network for food Classifier
AI Food detector; A model of Generative adversarial network for food Classifier
 
Yoga pose detection using deep learning project PPT.pptx
Yoga pose detection using deep learning project PPT.pptxYoga pose detection using deep learning project PPT.pptx
Yoga pose detection using deep learning project PPT.pptx
 
"Semantic Indexing of Wearable Camera Images: Kids’Cam Concepts"
"Semantic Indexing of Wearable Camera Images: Kids’Cam Concepts""Semantic Indexing of Wearable Camera Images: Kids’Cam Concepts"
"Semantic Indexing of Wearable Camera Images: Kids’Cam Concepts"
 
DSC Cloud Study Jams
DSC Cloud Study JamsDSC Cloud Study Jams
DSC Cloud Study Jams
 
DreamPose: Fashion Image to Video Synthesis via Stable Diffusion
DreamPose: Fashion Image to Video Synthesis via Stable DiffusionDreamPose: Fashion Image to Video Synthesis via Stable Diffusion
DreamPose: Fashion Image to Video Synthesis via Stable Diffusion
 
Object detection presentation
Object detection presentationObject detection presentation
Object detection presentation
 
Object recognition
Object recognitionObject recognition
Object recognition
 
Automated_attendance_system_project.pptx
Automated_attendance_system_project.pptxAutomated_attendance_system_project.pptx
Automated_attendance_system_project.pptx
 
Watson Visual Recognition - NYC JavaSig
Watson Visual Recognition - NYC JavaSigWatson Visual Recognition - NYC JavaSig
Watson Visual Recognition - NYC JavaSig
 
Introduction talk to Computer Vision
Introduction talk to Computer Vision Introduction talk to Computer Vision
Introduction talk to Computer Vision
 
502021435-12345678Minor-Project-Ppt.pptx
502021435-12345678Minor-Project-Ppt.pptx502021435-12345678Minor-Project-Ppt.pptx
502021435-12345678Minor-Project-Ppt.pptx
 
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine LearningMakine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
 
Easy path to machine learning (Spring 2020)
Easy path to machine learning (Spring 2020)Easy path to machine learning (Spring 2020)
Easy path to machine learning (Spring 2020)
 

Kürzlich hochgeladen

Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort serviceGurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort servicejennyeacort
 
Heart Disease Prediction using machine learning.pptx
Heart Disease Prediction using machine learning.pptxHeart Disease Prediction using machine learning.pptx
Heart Disease Prediction using machine learning.pptxPoojaBan
 
Oxy acetylene welding presentation note.
Oxy acetylene welding presentation note.Oxy acetylene welding presentation note.
Oxy acetylene welding presentation note.eptoze12
 
Gfe Mayur Vihar Call Girls Service WhatsApp -> 9999965857 Available 24x7 ^ De...
Gfe Mayur Vihar Call Girls Service WhatsApp -> 9999965857 Available 24x7 ^ De...Gfe Mayur Vihar Call Girls Service WhatsApp -> 9999965857 Available 24x7 ^ De...
Gfe Mayur Vihar Call Girls Service WhatsApp -> 9999965857 Available 24x7 ^ De...srsj9000
 
Past, Present and Future of Generative AI
Past, Present and Future of Generative AIPast, Present and Future of Generative AI
Past, Present and Future of Generative AIabhishek36461
 
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdfCCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdfAsst.prof M.Gokilavani
 
DATA ANALYTICS PPT definition usage example
DATA ANALYTICS PPT definition usage exampleDATA ANALYTICS PPT definition usage example
DATA ANALYTICS PPT definition usage examplePragyanshuParadkar1
 
What are the advantages and disadvantages of membrane structures.pptx
What are the advantages and disadvantages of membrane structures.pptxWhat are the advantages and disadvantages of membrane structures.pptx
What are the advantages and disadvantages of membrane structures.pptxwendy cai
 
Electronically Controlled suspensions system .pdf
Electronically Controlled suspensions system .pdfElectronically Controlled suspensions system .pdf
Electronically Controlled suspensions system .pdfme23b1001
 
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptxDecoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptxJoão Esperancinha
 
EduAI - E learning Platform integrated with AI
EduAI - E learning Platform integrated with AIEduAI - E learning Platform integrated with AI
EduAI - E learning Platform integrated with AIkoyaldeepu123
 
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube ExchangerStudy on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube ExchangerAnamika Sarkar
 
GDSC ASEB Gen AI study jams presentation
GDSC ASEB Gen AI study jams presentationGDSC ASEB Gen AI study jams presentation
GDSC ASEB Gen AI study jams presentationGDSCAESB
 
IVE Industry Focused Event - Defence Sector 2024
IVE Industry Focused Event - Defence Sector 2024IVE Industry Focused Event - Defence Sector 2024
IVE Industry Focused Event - Defence Sector 2024Mark Billinghurst
 
Risk Assessment For Installation of Drainage Pipes.pdf
Risk Assessment For Installation of Drainage Pipes.pdfRisk Assessment For Installation of Drainage Pipes.pdf
Risk Assessment For Installation of Drainage Pipes.pdfROCENODodongVILLACER
 
Work Experience-Dalton Park.pptxfvvvvvvv
Work Experience-Dalton Park.pptxfvvvvvvvWork Experience-Dalton Park.pptxfvvvvvvv
Work Experience-Dalton Park.pptxfvvvvvvvLewisJB
 

Kürzlich hochgeladen (20)

Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort serviceGurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
 
young call girls in Rajiv Chowk🔝 9953056974 🔝 Delhi escort Service
young call girls in Rajiv Chowk🔝 9953056974 🔝 Delhi escort Serviceyoung call girls in Rajiv Chowk🔝 9953056974 🔝 Delhi escort Service
young call girls in Rajiv Chowk🔝 9953056974 🔝 Delhi escort Service
 
Heart Disease Prediction using machine learning.pptx
Heart Disease Prediction using machine learning.pptxHeart Disease Prediction using machine learning.pptx
Heart Disease Prediction using machine learning.pptx
 
Oxy acetylene welding presentation note.
Oxy acetylene welding presentation note.Oxy acetylene welding presentation note.
Oxy acetylene welding presentation note.
 
Gfe Mayur Vihar Call Girls Service WhatsApp -> 9999965857 Available 24x7 ^ De...
Gfe Mayur Vihar Call Girls Service WhatsApp -> 9999965857 Available 24x7 ^ De...Gfe Mayur Vihar Call Girls Service WhatsApp -> 9999965857 Available 24x7 ^ De...
Gfe Mayur Vihar Call Girls Service WhatsApp -> 9999965857 Available 24x7 ^ De...
 
Past, Present and Future of Generative AI
Past, Present and Future of Generative AIPast, Present and Future of Generative AI
Past, Present and Future of Generative AI
 
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdfCCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
 
DATA ANALYTICS PPT definition usage example
DATA ANALYTICS PPT definition usage exampleDATA ANALYTICS PPT definition usage example
DATA ANALYTICS PPT definition usage example
 
What are the advantages and disadvantages of membrane structures.pptx
What are the advantages and disadvantages of membrane structures.pptxWhat are the advantages and disadvantages of membrane structures.pptx
What are the advantages and disadvantages of membrane structures.pptx
 
Electronically Controlled suspensions system .pdf
Electronically Controlled suspensions system .pdfElectronically Controlled suspensions system .pdf
Electronically Controlled suspensions system .pdf
 
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptxDecoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
 
EduAI - E learning Platform integrated with AI
EduAI - E learning Platform integrated with AIEduAI - E learning Platform integrated with AI
EduAI - E learning Platform integrated with AI
 
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube ExchangerStudy on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
 
GDSC ASEB Gen AI study jams presentation
GDSC ASEB Gen AI study jams presentationGDSC ASEB Gen AI study jams presentation
GDSC ASEB Gen AI study jams presentation
 
IVE Industry Focused Event - Defence Sector 2024
IVE Industry Focused Event - Defence Sector 2024IVE Industry Focused Event - Defence Sector 2024
IVE Industry Focused Event - Defence Sector 2024
 
Risk Assessment For Installation of Drainage Pipes.pdf
Risk Assessment For Installation of Drainage Pipes.pdfRisk Assessment For Installation of Drainage Pipes.pdf
Risk Assessment For Installation of Drainage Pipes.pdf
 
Work Experience-Dalton Park.pptxfvvvvvvv
Work Experience-Dalton Park.pptxfvvvvvvvWork Experience-Dalton Park.pptxfvvvvvvv
Work Experience-Dalton Park.pptxfvvvvvvv
 
Exploring_Network_Security_with_JA3_by_Rakesh Seal.pptx
Exploring_Network_Security_with_JA3_by_Rakesh Seal.pptxExploring_Network_Security_with_JA3_by_Rakesh Seal.pptx
Exploring_Network_Security_with_JA3_by_Rakesh Seal.pptx
 
POWER SYSTEMS-1 Complete notes examples
POWER SYSTEMS-1 Complete notes  examplesPOWER SYSTEMS-1 Complete notes  examples
POWER SYSTEMS-1 Complete notes examples
 
young call girls in Green Park🔝 9953056974 🔝 escort Service
young call girls in Green Park🔝 9953056974 🔝 escort Serviceyoung call girls in Green Park🔝 9953056974 🔝 escort Service
young call girls in Green Park🔝 9953056974 🔝 escort Service
 

Image-Based Ad Generation Using Deep Learning Object Detection

  • 1. AdWorld : Ad’s Based on Image Chetan Kulkarni San Jose State University Deep Learning
  • 2. Introduction • With the world getting completely online, Advertisers must innovate to increase the sales of product. • The idea at hand was using the continuous data (Image) and analyzing the image and producing ad’s
  • 3. Dasetset Collection 1. Used Googles OpenImagesV6 Dataset 2. Leveraged openimages PyPi API to download dataset required for the object detection. 3. Date Composition was image data and its specific annotated data for that object
  • 4. Data Example Annotation in Pascal Format Image Every image will have its annotation xml file , these both will be used as data for transfer learning
  • 5. Data Example Annotation in Pascal Format Image
  • 6. Object Detection : Techniques • In this project I have tried to use various Pretrained models to perform Object Detection and on that Generating Ad’s. • RetinaNet • Yolov3 • YoloV3 tiny • Transfer Learning on Yolov3
  • 7. RetinaNet object Detection Model built on COCO dataset, and supported following image detection:
  • 10. Now Need to Generate Ad’s Object Object Category Advertisement The object will be detected by models Admin will make Object Category, Each object will be mapped to one or more categories Ad’s will be made to target specific category
  • 11. Django Application • TV is a object • TV is connected to utilities category • Subwoofer Ad is targeted to utilities category • So whenever we have TV, we show all the Ad’s of utilities category
  • 12. • Model Detected TV in the image • We have Advertisements of Subwoofer and couch connected to TV • Hence The Ad’s
  • 13. Transfer Learning • Using Existing Yolov3 pretrained model. Tried to transfer learn for 3 specific imageset of table, clothing and bottle. • Couldn’t Train model on basic google collab GPU’s • Tried Google Collab PRO with High Ram and High Gpu • It took 3000s for one epoch • Because of limited resource couldn’t train for more than 24 hours • Hit The Accuracy of 50% for these custom object detection
  • 17. Extending the detection model to Videos • After the Image detection and Ad Generation on Images. • Tried to Generate Ad’s on the Videos • It Takes lot of processing and time. • Splitting Videos into 20 Frames per image, and running image detection model on them • And on overall Image detected, Performing Ad Generation
  • 20. Conclusion • Using Pretrained Models and Transfer learning we can take image detection to next level • Application of this image and object detection can be used in different places like traffic • Personalized Ads can be Generated based on specific user and Ad’s can be served real time with live video feed in Mall’s