Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

KPNDepth: Depth Estimation of Lane Images under Complex Rainy Environment

Zhengxu Shi

TL;DR

A novel dual-layer convolutional kernel prediction network for lane depth estimation in rainy environments that predicts two sets of kernels to mitigate depth loss and rain streak artifacts is introduced.

Abstract

Recent advancements in deep neural networks have improved depth estimation in clear, daytime driving scenarios. However, existing methods struggle with rainy conditions due to rain streaks and fog, which distort depth estimation. This paper introduces a novel dual-layer convolutional kernel prediction network for lane depth estimation in rainy environments. It predicts two sets of kernels to mitigate depth loss and rain streak artifacts. To address the scarcity of real rainy lane data, an image synthesis algorithm, RCFLane, is presented, creating a synthetic dataset called RainKITTI. Experiments show the framework's effectiveness in complex rainy conditions.

KPNDepth: Depth Estimation of Lane Images under Complex Rainy Environment

TL;DR

A novel dual-layer convolutional kernel prediction network for lane depth estimation in rainy environments that predicts two sets of kernels to mitigate depth loss and rain streak artifacts is introduced.

Abstract

Recent advancements in deep neural networks have improved depth estimation in clear, daytime driving scenarios. However, existing methods struggle with rainy conditions due to rain streaks and fog, which distort depth estimation. This paper introduces a novel dual-layer convolutional kernel prediction network for lane depth estimation in rainy environments. It predicts two sets of kernels to mitigate depth loss and rain streak artifacts. To address the scarcity of real rainy lane data, an image synthesis algorithm, RCFLane, is presented, creating a synthetic dataset called RainKITTI. Experiments show the framework's effectiveness in complex rainy conditions.
Paper Structure (17 sections, 9 equations, 6 figures, 3 tables)

This paper contains 17 sections, 9 equations, 6 figures, 3 tables.

Table of Contents

  1. Introduction
  2. PROPOSED METHOD
  3. RCFLane: Building Real Lane Data in Rainy Cloudy and Foggy Days
  4. Rain Layer
  5. Mask Layer
  6. Transmission Layer
  7. Dual-layer Pixel-wise Convolution Kernel Prediction Network
  8. Added Depth Estimation Model of DLKPN
  9. EXPERIMENTS
  10. Datasets
  11. Experimental Setup
  12. Metrics
  13. Results
  14. Image Reconstruction on RainKITTI
  15. Depth Estimation of Lane in Rainy Days
  16. ...and 2 more sections

Figures (6)

  • Figure 1: RCFLane Algorithm Implementation Example.
  • Figure 2: Example of Intermediate Image Generated by RCFLane Algorithm.
  • Figure 3: Display of Prediction Results of Single-layer KPN on RainKITTI.
  • Figure 4: DLKPN Structure Diagram.
  • Figure 5: The Overall Structure of The KPNDepth Framework.
  • ...and 1 more figures