GraphDepth integrates multi-scale GraphSAGE layers into a ResNet-101 U-Net with attention-gated skips and uncertainty estimation to deliver competitive monocular depth accuracy at lower computational cost than transformers.
What uncertainties do we needinBayesiandeeplearningforcomputervision?
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 Hybrid CNN-GNN Architecture for Monocular Depth Estimation
GraphDepth integrates multi-scale GraphSAGE layers into a ResNet-101 U-Net with attention-gated skips and uncertainty estimation to deliver competitive monocular depth accuracy at lower computational cost than transformers.