piergiaj Goto Github PK

activitynet

This repository is intended to host tools and demos for ActivityNet

avid

AViD Dataset: Anonymized Videos from Diverse Countries

bakeoff

blocks

A Theano framework for building and training neural networks

caffe

Caffe: a fast open framework for deep learning.

caffe-deconvnet

A deconvolutional network in caffe

caffe-to-theano

Convert a Caffe Model to a Theano Model

decision-net

Decision Net Research

generate-image

My veresion of "Generating Images from Captions with Attention"

gpu_flow

GPU based optical flow extraction in OpenCV

kinetics-i3d

Convolutional neural network model for video classification trained on the Kinetics dataset.

latent-subevents

This repository contains the code for our AAAI 2017 paper, "Learning Latent Sub-events in Activity Videos Using Temporal Attention Filters"

lintel

A Python module to decode video frames directly, using the FFmpeg C API.

mlb-youtube

MLB-YouTube dataset, code and models for fine-grained activity recognition (CVsports 2018)

pytorch-i3d

pyvideoresearch

A repositsory of common methods, datasets, and tasks for video research

representation-flow-cvpr19

Code and models for our CVPR'19 paper "Representation Flow for Action Recognition"

super-events-cvpr18

Code for our CVPR 2018 paper "Learning Latent Super-Events to Detect Multiple Activities in Videos"

tgm-icml19

Code for our ICML 2019 paper "Temporal Gaussian Mixture Layer for Videos"