DIVE-Doc: Downscaling foundational Image Visual Encoder into hierarchical architecture for DocVQA
In the DocVQA context, current end-to-end models either use lightweight [...]
PRISM: Pruning for Rank-adaptive Interpretable Segmentation Model with Application to Historical Document Multiband Images
Multispectral (MS) imaging reveals latent content in historical documents by [...]
Best Paper Award at ICCV 2025 – VisionDocs Workshop
The Synchromedia Laboratory is pleased to recognize the achievements of [...]
Synchromedia Laboratory Welcomes Kathy Baig, CEO of ÉTS, for an Official Visit Highlighting Innovation and Research Excellence
On May 20, 2025, the Synchromedia Laboratory had the [...]