Reading list in Transformer

We are a team from KAUST Vision-CAIR group and focus on the Multi-modal representation learning.

This repo is aimed to collect all the recent popular Transformer paper, codes and learning resources with respect to the domains of Vision Transformer, NLP and multi-modal, etc.

Recent News

CVPR multi-modal papers are collected in here

The code of VisualGPT is open sourced. They can be found here

The code and paper of LeViT is open sourced. They can be found here

The paper MLP-Mixer: An all-MLP Architecture for Vision is availble here

The code and paper of MDTER is open sourced. They can be found here

The code and papper of RelTransformer is open sourced. They can be found here

The code and paper of Twins-SVT is open sourced. They can be found here

Vision Transformer for deepfake detection. They can be found here

The code of VideoGPT is open sourced. They can be found here

The code of CoaT is open sourced. They can be found here

The code of Kaleido-BERT is open sourced. They can be found here

The code of TimeSformer is open sourced. They can be found here

The code of SwinTransformer is open sourced. They can be found here

Topics (paper and code)

• Image Transformer

• Video Transformer

• Video & Language & other modality Transformer

• Image & language & other modlity Trasformer

• Natural Language Processing Transformer

• Efficient Transformer

• MLP for Image Classification

• other interested papers in related domains

Review Paper in multi-modal

• Video-language

Tutorials and workshop

• Cross-View and Cross-Modal Visual Geo-Localization: IEEE CVPR 2021 Tutorial

• From VQA to VLN: Recent Advances in Vision-and-Language Research: IEEE CVPR 2021 Tutorial

Datasets

• Multi-modal Datasets

Blogs

• Lil's blogs

Tools

• PyTorchVideo a deep learning library for video understanding research

• horovod a tool for multi-gpu parallel processing

• accelerate an easy API for mixed precision and any kind of distributed computing

• hyperparameter search: optuna

• AI Conference Deadlines