/Computer-Vision

Computer-Vision

Image Formation

Image Processing

Model fitting and optimization

Feature detection and matching

Image alignment and stitching

Computational photography

Motion estimation

Structure from motion and SLAM

  • Geometric calibration
  • Stereo geometry with calibrated cameras
  • Stereo geometry with uncalibrated cameras
  • Multi-view structure from motion (MV-SFM)
  • SLAM (Simultaneous Localization And Mapping)

Depth estimation

  • Depth with stereo
    • Local stereo matching : Window matching, Adaptive support region
    • Semi-global stereo matching : Dynamic programming
    • Global stereo matching : Belief propagation, Graph cut

3D reconstruction

  • Shape from X
    • Shape from shading
    • Shape from photometric stereo
    • Shape from texture
    • Shape from structured lightening
  • Volumetric scene modeling
    • Model-based human reconstruction : Face(3DMM:3D morphable face model), Body(SMPL:Skinned multi-person linear model)

Image-based rendering

Deep learning

  • Supervised learning
    • KNN: K-Nearest Neighbors
    • Bayesian classification
    • Logistic regression
    • Linear classifier
    • Support vector machine
    • Decision trees and random forests
  • Unsupervised learning
    • Clustering : K-means clustering, GMM : Gaussian Mixture Model, Mean-shift clustering, etc.
    • PCA : Principal Component Analysis

Recognition