https://github.com/imsubst4nce/project-stereo
A project based on the use of stereoscopic images for computer vision related tasks.
https://github.com/imsubst4nce/project-stereo
computer-vision deep-learning depth-estimation disparity-map edge-detection monocular-depth-estimation stereo-vision
Last synced: about 1 year ago
JSON representation
A project based on the use of stereoscopic images for computer vision related tasks.
- Host: GitHub
- URL: https://github.com/imsubst4nce/project-stereo
- Owner: imsubst4nce
- Created: 2023-05-17T20:25:22.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2023-05-17T21:27:11.000Z (about 3 years ago)
- Last Synced: 2025-02-10T07:26:41.091Z (over 1 year ago)
- Topics: computer-vision, deep-learning, depth-estimation, disparity-map, edge-detection, monocular-depth-estimation, stereo-vision
- Language: Jupyter Notebook
- Homepage:
- Size: 14.2 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# project_stereo
**A project based on the use of stereoscopic images for computer vision related tasks.**
This project was implemented by me for the class of **Introduction to Image Processing - 4TTV414U** the **16/05/2023**.
The notebook contains complete documentation about the subject of the project, the methods and the tools used.
# Project Parts
1. Given 2 stereoscopic images(left/right), create 2 disparity maps using Block Matching algorithms(StereoBM and StereoSGBM).
2. Using the left image create 2 different depth maps using DenseDepth and MiDaS pretrained models.
3. Use 2 high-pass filters(Laplacian, Sobel) to test the previously generated disparity and depth maps in terms of edge-detection.
Disclaimer: I do not own the stereo photos
[Link](https://vision.middlebury.edu/stereo/data/scenes2014/)
> Special thanks to the creators of [DenseDepth](https://github.com/ialhashim/DenseDepth.git) and [MiDaS](https://github.com/isl-org/MiDaS.git)