Hello Viewers, in this video, a multi-class object classification problem using HOG features is explained. To demonstrate the implementation, simple geometrical shapes (Circle, Square, Star and Triangle) are taken for classification. As a classifier, ECOC (Error Correcting Output Codes) based multi-class SVM is used. The shapes image database is obtained from Kaggle.
The HOG feature is very popular and widely used for object detection in images. To understand the HOG feature computation, viewers are requested to watch my previous video of HOG feature computation.
This video includes following contents:
- Introduction.
- Proposed scheme for object Classification.
- Image Database Preparation.
- ECOC based Multi-Class SVM.
- Appropriate Cell Size selection for HOG feature.
- MATLAB Code for Shapes Classification (Multi-Class).
- MATLAB Code for Discrete Testing.
1. Link for previous video on HOG feature computation: Click Here
2. Link for Kaggle Dataset: https://www.kaggle.com/smeschke/four-shapes
3. Link to download original paper of N. Dalal and Bill Triggs:
https://lear.inrialpes.fr/people/triggs/pubs/Dalal-cvpr05.pdf
Download Resources:
1. Image Dataset (Modified): Download (Source: Kaggle)
2. Distorted Test Images: Download
