Related Experiment Video
Updated: Dec 8, 2025
07:05
Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
Published on: October 27, 2016
9.5K
Unsupervised Monocular Depth Estimation Method Based on Uncertainty Analysis and Retinex Algorithm
Chuanxue Song1, Chunyang Qi1, Shixin Song2
1College of Automotive Engineering, Jilin University, Changchun 130022, China.
Sensors (Basel, Switzerland)
|September 24, 2020
Summary
This study introduces novel unsupervised monocular depth estimation techniques. The methods improve reliability and handle moving objects, achieving competitive results in computer vision tasks.
Area of Science:
- Computer Vision
- 3D Scene Reconstruction
- Augmented Reality
Background:
- Unsupervised monocular depth estimation is crucial for 3D computer vision applications.
- Existing methods struggle with output reliability and dynamic scenes.
Discussion:
- A novel monocular depth estimation method leverages uncertainty analysis to quantify output reliability.
- A Retinex-based photometric loss function addresses issues caused by moving objects.
Key Insights:
- The proposed uncertainty analysis enhances neural network reliability in depth estimation.
- The Retinex-based loss function improves accuracy by mitigating moving object artifacts.
Outlook:
- Further research can explore integrating these techniques into real-time AR systems.
- Advancements may lead to more robust 3D reconstruction from single images.
More Related Videos
Related Concept Videos
01:15Depth Perception and Spatial Vision
1.5K
Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.
1.5K
00:59Uncertainty: Overview
1.4K
In analytical chemistry, we often perform repetitive measurements to detect and minimize inaccuracies caused by both determinate and indeterminate errors. Despite the cares we take, the presence of random errors means that repeated measurements almost never have exactly the same magnitude. The collective difference between these measurements - observed values - and the estimated or expected value is called uncertainty. Uncertainty is conventionally written after the estimated or expected value.
1.4K