Publications

PhD Thesis

  • "Spatio-temporal Convolutional Neural Networks for Early Action and Gesture Recognition", William Mocaër. PhD Thesis: HAL

Journal

  • "Early gesture detection in untrimmed streams: A controlled CTC approach for reliable decision-making", William Mocaër; Eric Anquetil; Richard Kulpa. Pattern Recognition (PR), 2024. Paper: HAL, Springer

Conferences

  • "Classifying the Unknown: In-Context Learning for Open-Vocabulary Text and Symbol Recognition", Tom Simon; William Mocaër; Pierrick Tranouez; Clement Chatelain; Thierry Paquet. 2025 19th International Conference on Document Analysis and Recognition (ICDAR), Wuhan, China. Paper: ArXiv.
  • "Early Recognition of Untrimmed Handwritten Gestures with Spatio-Temporal 3D CNN", William Mocaër; Eric Anquetil; Richard Kulpa. 2022 26th International Conference on Pattern Recognition (ICPR), Montreal, Canada. Paper: IEEE, HAL. Poster: PDF.
  • "3D Spatio-Temporal Convolutional Network for Early Recognition of Unsegmented Handwritten Gestures", William Mocaër; Eric Anquetil; Richard Kulpa. 2022 Reconnaissance des Formes, Image, Apprentissage et Perception (RFIAP), Vannes, France. Paper: HAL.
  • "Online Spatio-temporal 3D Convolutional Neural Network for Early Recognition of Handwritten Gestures", William Mocaër; Eric Anquetil; Richard Kulpa. 2021 International Conference on Document Analysis and Recognition (ICDAR), Lausanne, Switzerland. Paper: Springer, HAL.

In Progress

  • "Tiramisu: Tiered Transformers for Hierarchical Structure Understanding", William Mocaër; Clément Chatelain; Stéphane Nicolas; Thierry Paquet; Tom Simon; Pierrick Tranouez

Posters