EECS 504 - Computer Vision

This page contains my course work from EEECS 504 (Fall 2022)

Programming language: Python
Frameworks / Library: Pytorch, SciPy
Topics:
- Linear clssifiers
- Gradient descent: Batch gradient descent, Stochastic gradient descent, Gradient descent with momentum, Gradient descent with RMSProp, Adam
- Fully-connected networks
- Convolutional networks: Alexnet, VGG, ResNet, Encoder-decoder architecture,
- Recurrent networks: RNN, LSTMs(Long Short Term Memory)
- Attention and transformers: Self-attention, Multi-head attention, Transformer, ViT(Vision Transformer)
- Object detection: R-CNN, Fast R-CNN, Faster R-CNN, YOLO, FCOS,
- Image segmentation
- Video classification: 3D CNN
- Generative models: GANs, VAEs(Variational AutoEncoders), Autoregressive models, Diffusion, Vector-quantized VAEs
- Recent architectures: NeRF(Neural Radience Field), BERT(Bidirectional Encoder Representations from Transformers), Image GPT, DETR(Detection Transformer)

Homeworks

1. Pet Edge Detection

Topic: Edge detection, Convolution

2. Signal Processing

Topic: Image blending, Fourier transform, Image compression

3. Introduction to Machine Learning

Topic: KNN, Multinomial logistic classification

4. Backpropagation

Topics: Backpropagation, Multi-layer perceptron

5. Scene Recognition

Topics: MiniVGG, MiniVGG-BN, Residual Networks

6. Image Synthesis

Topics: pix2pix, conditional GAN, U-net, receptive field, style transfer

7-1. Object Detection

Topics: Object detection, FCOS, FPN(Feature Pyramid Network), GIoU(Generalized Intersection-over-Union), BCE(Binary Cross Entropy), mAP(mean Average Precision)