TakuNet goes journal: Image and Vision Computing

New Publication! The extended version of TakuNet, “TakuNet: Energy-efficient models for real-time aerial disaster response and monitoring on edge devices”, is now available online in Image and Vision Computing (Elsevier), vol. 175, art. 106151.

The journal version broadens the original WACVW 2025 work into a family of energy-efficient models, with a deeper study of the accuracy/energy trade-off on embedded accelerators for aerial disaster response and monitoring.

Authors: Daniel Rossi, Gianluca Filippini, Andrea Torlai, Guido Borghi, Roberto Vezzani Code: github.com/DanielRossi1/TakuNetV2