UniFormerV2
Video network builder
A software framework for building powerful video networks by augmenting pre-trained image vision transformers with efficient temporal learning models.
[ICCV2023] UniFormerV2: Spatiotemporal Learning by Arming Image ViTs with Video UniFormer
7 stars
2 watching
2 forks
Language: Jupyter Notebook
last commit: almost 3 years agoLinked from 1 awesome list
amrzv/awesome-colab-notebooksRelated projects:
| Repository | Description | Stars |
|---|---|---|
 jay-mahadeokar/pynetbuilder | A modular Python framework for building and generating neural networks using Caffe's NetSpec class. | 328 |
 ahmedfgad/numpycnn | A Python implementation of a Convolutional Neural Network from scratch using NumPy for building CNNs from scratch | 577 |
 opengvlab/internvideo | Develops general video foundation models and related datasets for multimodal understanding and generation through generative and discriminative learning. | 1,467 |
 randl/mobilenetv2-pytorch | An implementation of MobileNetV2 in PyTorch for image classification tasks. | 271 |
 singularity42/vgan-tensorflow | An implementation of a deep learning model to generate videos with dynamic scenes | 15 |
 ternaus/ternausnetv2 | A deep learning model for automatic instance segmentation of building footprints from satellite imagery | 548 |
 researchmm/sttn | Proposes a deep learning model to fill missing regions in video frames and generate completed videos | 480 |
 nvidia/unsupervised-video-interpolation | This project provides a framework for unsupervised video interpolation using cycle consistency. | 107 |
 neuleaf/mobilenetv2 | An implementation of MobileNet V2 using TensorFlow, providing basic functionality for training and testing the network | 150 |
 yitu-opensource/t2t-vit | A deep learning framework for training vision transformers from scratch on image data. | 1,162 |
 p-christ/nn_builder | Builds neural networks with less boilerplate code by providing a standardized interface for different architectures | 166 |
 ibm/max-sports-video-classifier | This project provides a pre-trained video classification model that categorizes sports videos into their respective sports. | 23 |
 nus-hpc-ai-lab/videosys | A comprehensive toolkit for high-performance video generation and processing | 1,819 |
| ibm/max-inception-resnet-v2 | An image classification model using a third-generation deep residual network. | 27 |
 timvn/gmnest | A socket.IO-based extension for Game Maker, enabling real-time communication and networking capabilities. | 1 |