fourierer/Video_Classification_ResNet3D_R2plus1D_ip-CSN_train-UCF101-HMDB51-Kinetics400-from-scratch

Stars
25
Rank 952,485 (Top 19 %)
Language
Python
Created about 4 years ago
Updated about 4 years ago

fourierer/Video_Classification_ResNet3D_R2plus1D_ip-CSN_train-UCF101-HMDB51-Kinetics400-from-scratch

fourierer

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Using ResNet3D-50,R(2+1)D-50, and ip_CSN-50 to train UCD-101,HMDB-51 and Kinetics-400 from scratch.

Video_Classification_ResNet3D_Pytorch

Using ResNet3D to train on Kinetics form scratch or fine-tune on UCF-101(or others) with Kinetics pretrained model.

Knowledge_Base

collect some questions in AI interview, include mathematics, computer algorithm, dl or ml model and so on

Learning_Papers_of_Human_Pose_Estimation

Learning SimpleBaseline and LightweightBaseling for Human Pose Estimation

ffmpeg

Just for learning ffmpeg

classification_of_ImageNet_Resnet_Pytorch

classify the imagenet using resnet model in pytorch

Recognizing-Image-Style

Recognizing Image Style with Resnet or other model in pytorch

multiple_label_classification

multiple label classification with MobileNetV2, ShuffleNetV2 and GhostNet

Learn_GhostNet_pytorch

This repository is created for learning the GhostNet paper and code, the code copy from "https://github.com/iamhankai/ghostnet.pytorch"

Pedestrian_Attribute_Recognition

Recognizing pedestrian attributes on RAP(V2) via pytorch, copy from 'https://github.com/dangweili/pedestrian-attribute-recognition-pytorch' ,annotate the code and the learn model

Training_model_with_Data_Parallelism

Introduce several ways of data parallelism training, train model on one machine and seven GPUs.