Dense depth distillation with out-of-distribution simulated images
Lists
Only you see this
Abstract
Highlights•Novel data-free distillation for depth estimation using OOD simulated images.•Analysis of scene similarity, dataset size, and domain gap for accurate distillation.•Specialized framework addressing challenges in data-free depth distillation.•Superior performance with different depth estimation models and benchmark datasets.
BibTeX
@article{hu2022dense,
author = {Junjie Hu and Chenyou Fan and Mete Ozay and Hualie Jiang and Tin Lun Lam},
title = {{Dense depth distillation with out-of-distribution simulated images}},
journal = {Knowledge-Based Systems},
year = {2022},
}