Face recognition / Image Classification / Computer Vision
Budget: €30 – €250 EUR
In the first part of this assignment, you will start from a small dataset and construct/learn features that describe the data. We discriminate two types of features, handcrafted features and features learned from data. In this assignment you will experiment with Histogram of Oriented Gradients (HOG) or Scale Invariant Feature Transform (SIFT) feature descriptors. You will explore Principal Component Analysis (PCA) in the context of face detection, so-called eigenfaces. In the second part of the assignment you will use the constructed features for face recognition using your favorite classifier. In the third part of this assignment you will participate in a kaggle competition and attempt to improve your classification pipeline in any way you
please. Lastly, you will reflect on the entire assignment and write a final discussion, this is
on top of the discussion and comments that you already provided in the previous parts. In
summary, the assignment is broken down into four main parts:
1. build feature representations using handcrafted and non-handcrafted techniques
2. use and compare the feature representations in context of classification of faces
3. attempt to improve your model and participate in the kaggle competition
4. discussion
Price can be discussed.
please. Lastly, you will reflect on the entire assignment and write a final discussion, this is
on top of the discussion and comments that you already provided in the previous parts. In
summary, the assignment is broken down into four main parts:
1. build feature representations using handcrafted and non-handcrafted techniques
2. use and compare the feature representations in context of classification of faces
3. attempt to improve your model and participate in the kaggle competition
4. discussion
Price can be discussed.