Joan Puigcerver
Scholar

Joan Puigcerver

Google Scholar ID: ZHnRsrsAAAAJ
Google DeepMind
Machine LearningDeep LearningComputer VisionHandwriting Recognition
Citations & Impact
All-time
Citations
7,967
 
H-index
26
 
i10-index
37
 
Publications
20
 
Co-authors
10
list available
Publications
20 items
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Research Experience
  • Fall 2018–Present: Software Engineer in Research, Google Brain, Zurich
  • Summer/Fall 2017: Software Engineer Intern, Google Research (Video Content Analysis team), applied deep learning to content-based recommendation for newly uploaded videos to address the 'fresh start' problem
  • Summer/Fall 2015: Intern, Google Geo (StreetSmart team), improved business change detection from Street View using CNNs and LSTMs to keep Google Maps listings fresh at lower cost
  • Summer 2014: Computer Vision Engineer, Blinkfire Analytics, built backend for brand logo detection on social media images using multiple classification models, enhancing speed and accuracy
  • Summer 2013: Intern, Google Research (OCR team), developed a synthetic data generation framework for OCR; trained state-of-the-art models for Google Books, Ads OCR, and in-the-wild OCR using only synthetic data
  • Summer 2012: Intern, Google Research (OCR team), developed one of the first deep learning-based feature extraction systems, significantly reducing character error rates across 40+ languages for Google Books