Computer Vision Model Implementation by Pytorch

This is the repo including multiple Computer Vision Model Implementation with full explanation. For each model, there is tutorial for any operation.

Model with task category:

Image Classification

1. ResNet
2. Vision Transformer
3. Swin Transformer

2D Object Detection

1. SSD
2. CenterNet
3. DETR (todo)
4. RT-DETR (todo)
5. Deformable Convolutional Networks (todo)
6. Deformable-DETR (todo)

2D Semantic Segmentation

1. DeepLabV3
2. DeepLabV3+
3. SegFormer (todo)
4. Segmenter (todo)
5. OCRNet (todo)

3D Semantic Segmentation/Classification

1. PointNet
2. PointNet++
3. PointTransformer (todo)

3D Object Detection

1. VoteNet (todo)
2. BevFormer (todo)
3. PointPillar (todo)
4. VoxelNet (todo)

GenAI

Diffusion Model

1. Denoising Diffusion Probability Model (DDPM)
2. Denoising Diffusion Implicit Model (DDIM)
3. Classifier-Free Diffusion Guidance (CFDG)
4. Image Super-Resolution via Iterative Refinement (SR3) (todo)

Auto Encoder

1. Variational Autoencoder (VAE)
2. Vector Quantization Variational Autoencoder (VQVAE)

Pixel CNN

1. PixelCNN
2. Gated PixelCNN

GAN

1. CGAN
2. DCGAN