MonoSAOD enables robust monocular 3D object detection under sparse supervision by combining geometry-preserving road patch augmentation with prototype-guided pseudo-label filtering.
Co- mining: Self-supervised learning for sparsely annotated object detection
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
-
MonoSAOD: Monocular 3D Object Detection with Sparsely Annotated Label
MonoSAOD enables robust monocular 3D object detection under sparse supervision by combining geometry-preserving road patch augmentation with prototype-guided pseudo-label filtering.