Skip to content

Latest commit

 

History

History
60 lines (51 loc) · 2.35 KB

README.md

File metadata and controls

60 lines (51 loc) · 2.35 KB
Raw

Advanced Computer Vision: Homework

Contents

Project 1: Image Matching

  • keywords: Block Matching

Project 2: Optical Flow

  • keywords: Horn–Schunck, Lucas-Kanade

Project 3: Stereo Reconstruction

  • keywords: stereo camera, image rectification, disparity map, point clouds

Final Project

  • Title: Improved Real-Time Pipeline for 3D Mesh Reconstruction from RGB-D Cameras
  • external link: abstract, video

Note

  • References are listed at the bottom of this page.
  • The pdf file MyNotes.pdf includes the material for preparing my midterm and final exams in the class of Advanced Computer Vision. These notes are provided for personal use only and may not be used for commercial purposes. If you notice any typos or have inquiries regarding the use of these notes, feel free to contact me. Thank you!

Project 1: Image Matching

Python Usage

python3 image_matching.py

animation_motion_with_block

Project 2: Optical Flow

Python Usage

python3 optical_flow.py

Results from Horn–Schunck optical flow:

sphere_hs_flow table_hs_flow

Results from Lucas-Kanade optical flow:

sphere_lk_flow

Project 3: Stereo Reconstruction

Python Usage

python3 stereo_reconstruction.py

Result (Demonstration Video)

stereo_reconstruction_open3d

Note

  • All output files are saved in the results folder.
  • Some projects are modified from the class of Advanced Computer Vision at National Taiwan University.

Resources and References

[1] Lecture Notes from Chiou-Shann Fuh's class of Advanced Computer Vision (Spring 2023)
[2] optical flow from lmiz100's github
[3] stereo algorithm from sushlokshah's github
[4] stereo datasets