LiRCDepth shows that a MobileNetV2-based radar-camera depth estimator can recover much of the accuracy of a ResNet-based teacher via feature, structure, and uncertainty-weighted depth distillation.
Learning depth with convolutional spatial propagation network,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
LiRCDepth: Lightweight Radar-Camera Depth Estimation via Knowledge Distillation and Uncertainty Guidance
LiRCDepth shows that a MobileNetV2-based radar-camera depth estimator can recover much of the accuracy of a ResNet-based teacher via feature, structure, and uncertainty-weighted depth distillation.