wanglimin

dense_flow

OpenCV Implementation of different optical flow algorithms

ARTNet

Appearance-and-Relation Networks

UntrimmedNet

Weakly Supervised Action Recognition and Detection

MRCNN-Scene-Recognition

MR-CNNs for Large-Scale Scene Recognition

TDD

Trajectory-pooled Deep-Convolutional Descriptors

Places205-VGGNet

Places205-VGGNet models for scene recognition

Actionness-Estimation

Actionness Estimation Using Hybrid Fully Convolutional Networks

improved_trajectory

Improved trajectories for action recognition

wanglimin.github.io

Hompage of Wang Limin

Action-Recognition-Report

Publish report

action.sr_cnn

code for Two-Stream SR-CNNs for Action Recognition in Videos 2016

PRML

Pattern recognition and machine learning toolbox

a

C3D

C3D is a modified version of BVLC caffe to support 3D ConvNets.

CMPT

[IJCV 2021] Cross-Modal Pyramid Translation for RGB-D Scene Recognition

deep-residual-networks

Deep Residual Learning for Image Recognition

warp_flow