Creating an artwork finder with the Cloud Vision API

on YouTube ↗17m

" summary

Shows how to build an artwork finder using Google Cloud Vision: use Face Landmarks to crop thumbnails and reject images with multiple faces, and SafeSearch to block or blur explicit uploads. Uses Web Detection to retrieve higher‑resolution matches and pages that identify artists, OCR to extract signatures, and trains a custom classifier with AutoML Vision from images on a GCS bucket (uploads via gsutil); about 800 labeled images trains in ~10 minutes. Code is published on GitHub.

AI-written summary · about

" chapters (9)

1. 0:00Project intro
2. 1:09Face detection thumbnails
3. 4:35Filter explicit content
4. 5:57Higher quality images
5. 7:20Expand artwork collection
6. 8:42Find artwork artist
7. 11:06Content categorization
8. 12:53AutoML Vision categorization
9. 17:07Conclusion

" related talks

• 37mAnalyzing Images with Google’s Cloud Vision APISara Robinson · DevFestMN 2016
• 35mComputer vision: British ClayValentin Deleplace · DevFest Ukraine 2018
• 59mMachine Learning and Big Data Magic with Node.js and Google CloudBret McGowen, Sara Robinson · NDC Sydney 2017
• 57mBuilding smarter solutions with Machine Learning, from magic to realityLaurent Picard · code.talks commerce 2019
• 47mBuilding smarter apps with Machine Learning, from magic to realityLaurent Picard · GDG DevFest Poland 2018 in Wrocław
• 36mCloud AI Services: What they are and how to use themKarl Weinmeister · Global Artificial Intelligence Conference 2020
• 49mKeynote: From Zero to ML on Google Cloud PlatformSara Robinson · DevFestMN 2018
• 01h 04mBuilding and Training your own Custom Image Recognition AIDavid Giard · NDC Oslo 2018