An end-to-end neural transceiver jointly trained with 3D Gaussian Splatting loss enables sparse pilots and higher-fidelity large-scale 3D reconstructions from low-altitude aerial images.
AI and ML for NR air interface in Rel-20 5G-Advanced
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Efficient Transceiver Design for Aerial Image Transmission and Large-scale Scene Reconstruction
An end-to-end neural transceiver jointly trained with 3D Gaussian Splatting loss enables sparse pilots and higher-fidelity large-scale 3D reconstructions from low-altitude aerial images.