Upgrade to Pro — share decks privately, control downloads, hide ads and more …

โฐโฐโฐ ๐™ˆ๐™–๐™˜๐™๐™ž๐™ฃ๐™š ๐™‡๐™š๐™–๐™ง๐™ฃ๐™ž๐™ฃ๐™œ ๐™Ž๐™ช๐™ฅ๐™š๐™ง๐™ฅ๐™ค๐™ฌ๐™š๐™ง๐™จ ๐™ฅ๐™š๐™ง ๐™ก๐™– ๐™ฉ๐™ช๐™– ๐˜ผ๐™ฅ๐™ฅใƒป๐™๐™ช๐™ฉ๐™ฉ๐™ž ๐™– ๐™—๐™ค๐™ง๐™™๐™ค ๐™™๐™ž ๐™ƒ๐˜ฟ๐™‚ ๐™„๐™ฉ๐™–๐™ก๐™ž๐™–! โฑโฑโฑ

โฐโฐโฐ ๐™ˆ๐™–๐™˜๐™๐™ž๐™ฃ๐™š ๐™‡๐™š๐™–๐™ง๐™ฃ๐™ž๐™ฃ๐™œ ๐™Ž๐™ช๐™ฅ๐™š๐™ง๐™ฅ๐™ค๐™ฌ๐™š๐™ง๐™จ ๐™ฅ๐™š๐™ง ๐™ก๐™– ๐™ฉ๐™ช๐™– ๐˜ผ๐™ฅ๐™ฅใƒป๐™๐™ช๐™ฉ๐™ฉ๐™ž ๐™– ๐™—๐™ค๐™ง๐™™๐™ค ๐™™๐™ž ๐™ƒ๐˜ฟ๐™‚ ๐™„๐™ฉ๐™–๐™ก๐™ž๐™–!ย โฑโฑโฑ

Stanco del solito classificatore di immagini? Vuoi rendere la tua app piรน intelligente?
Non bisogna essere scienziati per donare alla tua App mobile i superpoteri del Machine Learning!
Scopriamo insieme come rendere l'esperienza utente "quasi magica" utilizzando i servizi offerti da HMS ML Kit.
Analizzeremo casi pratici e tanto codice che ti aiuteranno a capire cosa puoi fare con questi servizi innovativi.

Giovanni Laquidara

February 25, 2021
Tweet

More Decks by Giovanni Laquidara

Other Decks in Programming

Transcript

  1. Machine Learning Superpowers per la tua App

  2. Giovanni Laquidara https://speakerdeck.com/joaobiriba https://medium.com/@joaolaq https://medium.com/huawei-developers/solve-a-word-search-game-using-firebase-ml-kit- and-huawei-ml-kit-8b9cc06bf4ce Huawei Developer Advocate Software

    Engineer loving Android, VR/AR and ML
  3. 2011 https://a16z.com/2011/08/20/why-software-is-eating-the-world/ Software is eating the world!

  4. 2019 https://www.forbes.com/sites/cognitiveworld/2019/08/29/soft ware-ate-the-world-now-ai-is-eating-software/#337963295810 AI is Eating Software

  5. AI for Programmers

  6. Machine learning (ML) is the study of computer algorithms that

    improve automatically through experience.[1][2] It is seen as a subset of artificial intelligence. Machine learning algorithms build a mathematical model based on sample data, known as "training data", in order to make predictions or decisions without being explicitly programmed to do so.[3] Machine learning algorithms are used in a wide variety of applications, such as email filtering and computer vision, where it is difficult or infeasible to develop conventional algorithms to perform the needed tasks. https://en.wikipedia.org/wiki/Machine_learning Machine Learning
  7. Machine learning Machine Learning

  8. Mobile development + ML

  9. โ€ข Firebase ML / ML Kit โ€ข Amazon Amplify โ€ข

    Microsoft Azure โ€ข Well Known feature โ€ข Sometimes not familiar API โ€ข Configurations Common frameworks
  10. Huawei ML Kit โ€ข Innovative features โ€ข Familiar API โ€ข

    Not forgetting about the โ€œoldโ€ Java :D https://developer.huawei.com/consumer/en/hms/
  11. Build your AI application with ML KIT New vison and

    language AI experience ML Kit provides various visual and language machine learning service APIs. O b j e c t d e t e c t i o n a n d t r a c k i n g i m a g e s u p e r - r e s o l u t i o n I D c a r d / b a n k c a r d / g e n e r a l c a r d r e c o g n i t i o n T e x t r e c o g n i t i o n R e a l - t i m e s p e e c h r e c o g n i t i o n P r o d u c t V i s u a l s e a r c h I m a g e s e g m e n t a t i o n L a n d m a r k R e c o g n i t i o n Text D o c u m e n t r e c o g n i t i o n I m a g e c l a s s i f i c a t i o n Images Voice language T e x t T r a n s l a t i o n L a n g u a g e R e c o g n i t i o n s p e e c h s y n t h e s i s Reads text in different forms. Text OCR, including text, document, bank card, ID card, and general card OCR Provide rich and accurate visual services. Classify images, detect objects, track them, identify landmarks with their names, segment images with images of people, sky, and plants, and take photos to obtain the same or similar products. Machine learning services related to natural languages. Provides voice-to- text and text-to-speech services and the capability of translating between different languages. Face detection services with contours, key points, emotions, and attributes and precise bone recognition Face & Person F a c e d e t e c t i o n S k e l e t o n d e t e c t i o n H a n d k e y p o i n t i d e n t i f i c a t i o n L i v e n e s s d e t e c t i o n d i g i t a l h u m a n Custom Model C u s t o m i z i n g I m a g e C a t e g o r i e s C u s t o m i z i n g o b j e c t d e t e c t i o n C u s t o m i z i n g T e x t C a t e g o r i e s Using Huawei IDE, you can customize models with only a few training and test data sets. V i d e o C o u r s e C r e a t o r D o c u m e n t S k e w C o r r e c t i o n
  12. A l l m o d e l s c

    o v e r e d Cover all Android/iOS devices F r e e Free on the device side and limited on the cloud side S i m p l e i n t e g r a t i o n Simple and quick integration of ML Kit G l o b a l c o v e r a g e Fast and direct service worldwide โˆš Huawei โˆš Samsung โˆš Xiaomi โˆš OPPO iPhone iPad >= iOS 9.0 ๏ƒผ Faster response ๏ƒผ More secure data ๏ƒผ Free Unlimited Call ๏ƒผ NPU/CPU = Higher accuracy = Smaller packet size = More functions ๏ƒผ Europe ๏ƒผ Asia, Africa, and Latin America ๏ƒผ Russia Region ๏ƒผ China Region = High availability, ensuring continuous and stable services = High concurrency, high throughput, and high reliability = Low latency ๏ƒผ Complete service integration within 1 to 2 days ๏ƒผ Optional default interactive interface or custom interface = Demo github open source = Complete developer manual = Video guide Android device โˆš Others โˆš VIVO iOS device Device service Coverage area Easy integration Cloud-side services Service Provisioning Development Guide HMS ML Kit provides all Android/iOS models, NPU/CPU heterogeneous adaptation, and integrated AI openness capabilities for device-cloud synergy. Core Values of ML KIT
  13. Text-related Services โ€ขText recognition: Extracts text from images โ€ขDocument recognition:

    Recognizes text with paragraph formats โ€ขID card recognition: Recognizes text in images or camera streams of ID cards. โ€ขBank card recognition: Quickly recognizes information such as the bank card number, covering mainstream bank cards โ€ขGeneral card recognition: extract required information from any fixed-format cards โ€ขForm recognition: Recognize and return form structure information (including rows, columns, and coordinates of cells) from input images.
  14. Text-related Services https://github.com/HMS-Core/hms-ml-demo

  15. Bank Card Recognition

  16. Bank Card Recognition

  17. Bank Card Recognition

  18. Language/Voice-related Services โ€ข Real-time translation โ€ข On-device translation โ€ข Real-time

    language detection โ€ข On-device language detection โ€ข Automatic speech recognition โ€ข Text to speech โ€ข On-device text to speech โ€ข Audio file transcription โ€ข Real-time transcription โ€ข Sound detection
  19. Sound Detection

  20. Sound Detection

  21. Image-related Services โ€ข Image classification โ€ข Object detection and tracking

    โ€ข Landmark recognition โ€ข Image segmentation โ€ข Product visual search โ€ข Image super-resolution โ€ข Document skew correction โ€ข Text image super-resolution โ€ข Scene detection โ€ข Table recognition
  22. Object Recognition

  23. Image Segmentation

  24. Face/Body-related โ€ข Face detection โ€ข Skeleton detection โ€ข Liveness detection

    โ€ข Hand gesture recognition โ€ข Face verification
  25. ML Kit Hands Tracking / Gestures

  26. ML Kit Body pose

  27. Custom Model Quickly train and generate a model in your

    app, and package it with your app or upload it to the ML Kit model hosting platform Model conversion Model transfer learning
  28. HMS ML KIT Custom Model Local Model Hosted Model

  29. HMS ToolKit https://developer.huawei.com/consumer/en/huawei-toolkit/

  30. HMS ToolKit

  31. HMS ToolKit

  32. HMS ML KIT https://github.com/HMS-Core/hms-ml-demo

  33. HMS ML KIT https://github.com/joaobiriba/WordSearchSolver

  34. Show the code

  35. Show the code

  36. Dependencies

  37. HMS ML KIT https://github.com/joaobiriba/QuickFaceAnalyzer https://github.com/joaobiriba/QuickHandAnalyzer Sample using Android CameraX

  38. News media content production MOOC course development Ad content production

    F e a t u r e H M S Virtual image customization support Sound customization support multilingual broadcast support emoticon generation support Video format 4K video Dynamically changing the background support Digital Person
  39. Thanks! giovanni.laquidara@huawei.com