From df1decee41ad269661411b4326e8ded6158a575a Mon Sep 17 00:00:00 2001 From: Chuck Cho Date: Thu, 7 Sep 2017 13:13:41 -0400 Subject: [PATCH 1/2] fixed external links --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index dbaa057..05ba1e9 100644 --- a/README.md +++ b/README.md @@ -10,7 +10,7 @@ CVPR 2017 conference paper. "Quo Vadis" introduced a new architecture for video classification, the Inflated 3D Convnet or I3D. Here we release Inception-v1 I3D models trained on the -[Kinetics dataset](www.deepmind.com/kinetics) training split. +[Kinetics dataset](http://www.deepmind.com/kinetics) training split. In our paper, we reported state-of-the-art results on the UCF101 and HMDB51 datasets from fine-tuning these models. I3D models pre-trained on Kinetics also @@ -134,7 +134,7 @@ Two-Stream I3D | 74.2 / 91.3 | 71.6 / 90.0 ### Sample data and preprocessing -The release of the [DeepMind Kinetics dataset](www.deepmind.com/kinetics) only +The release of the [DeepMind Kinetics dataset](http://www.deepmind.com/kinetics) only included the YouTube IDs and the start and end times of the clips. For the sample data here, we use a video from the UCF101 dataset, for which all the videos are provided in full. The video used is `v_CricketShot_g04_c01.mp4` which From 2da266af20fa24b467ef31e9aa6e7f5bb2659b5f Mon Sep 17 00:00:00 2001 From: Chuck Cho Date: Thu, 7 Sep 2017 13:20:19 -0400 Subject: [PATCH 2/2] minor fix --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 05ba1e9..4961f7a 100644 --- a/README.md +++ b/README.md @@ -5,7 +5,7 @@ This repository contains trained models reported in the paper "[Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://arxiv.org/abs/1705.07750)" by Joao Carreira and Andrew -Zisserman. The paper was posted on arXiv in May 2017, and will be published as a +Zisserman. The paper was posted on arXiv in May 2017, and was published as a CVPR 2017 conference paper. "Quo Vadis" introduced a new architecture for video classification, the Inflated