A Few-Shot Learning Framework for Depth Completion Based on Self-Training with Noise and Pixel-Wise Knowledge Distillation

A Few-Shot Learning Framework for Depth Completion Based on Self-Training with Noise and Pixel-Wise Knowledge Distillation

Depth completion generates a comprehensive depth map by utilizing sparse depth data inputs, supplemented by guidance provided by an RGB image. Deep neural network models depend on annotated datasets for optimal training. However, when the quantity of training data is limited, the generalization capa...

Full description

Saved in:
Bibliographic Details
Main Authors: Shijie Zhang, Shengjie Zhao, Jin Zeng, Hao Deng
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Applied Sciences
Subjects:
depth completion
few-shot learning
knowledge distillation
self-training
Online Access:https://www.mdpi.com/2076-3417/15/9/4740
Tags: Add Tag
No Tags, Be the first to tag this record!

Internet

https://www.mdpi.com/2076-3417/15/9/4740

Similar Items