Self-Supervised Learning based Depth Estimation from Monocular Images

TL;DR

This paper explores enhancements to state-of-the-art deep learning models for monocular depth estimation, aiming to improve accuracy and generalization by integrating pose estimation, semantic segmentation, and data augmentation techniques.

Contribution

The paper proposes novel extensions to existing depth estimation models, including the integration of pose estimation, semantic segmentation, and weather augmentations for improved performance.

Findings

Potential performance improvements with proposed extensions

Enhanced generalization through weather augmentations

Framework for combining multiple techniques in depth estimation

Abstract

Depth Estimation has wide reaching applications in the field of Computer vision such as target tracking, augmented reality, and self-driving cars. The goal of Monocular Depth Estimation is to predict the depth map, given a 2D monocular RGB image as input. The traditional depth estimation methods are based on depth cues and used concepts like epipolar geometry. With the evolution of Convolutional Neural Networks, depth estimation has undergone tremendous strides. In this project, our aim is to explore possible extensions to existing SoTA Deep Learning based Depth Estimation Models and to see whether performance metrics could be further improved. In a broader sense, we are looking at the possibility of implementing Pose Estimation, Efficient Sub-Pixel Convolution Interpolation, Semantic Segmentation Estimation techniques to further enhance our proposed architecture and to provide fine-grained and more globally coherent depth map predictions. We also plan to do away with camera intrinsic parameters during training and apply weather augmentations to further generalize our model.