Kinetics-Data-Preprocessing

Overview

Kinetics-Data-Preprocessing

Kinetics-400 and Kinetics-600 are common video recognition datasets used by popular video understanding projects like SlowFast or PytorchVideo. However, their instruction of dataset preparation is too brief. Therefore, this project provides a more detailed instruction for Kinetics-400/-600 data preprocessing.

Download the raw videos

There are multiple ways to download the raw videos of Kinetics-400 and Kinetics-600. Here, I list two common choices that I found to be simple and fast:

  1. Download the videos via the official scripts. However, I noticed that this option is very slow, so I personally recommend the next choice.

  2. Download the compressed videos from the Common Visual Data Foundation Servers following the repository, which is much faster as they organized 650,000 independent video clips into several compressed files.

Resize the videos

The common data preprocessing of Kinetics requires all videos to be resized to the short edge size of 256. Therefore, I use the moviepy package to do so. The package can be easily installed by the following command:

pip install moviepy

Then, you can use the resize_video.py to resize all the videos within the given folder by following command:

python resize_video.py --size 256 --path YOUR_VIDEO_CONTAINER

IMPORTANT! Note that the resize_video.py will replace the original mp4 files. If you want to keep the original files, please make copys before resizing.

Prepare the csv annotation files

Following SlowFast, we also need to prepare the csv annotation files for training, validation, and testing set as train.csv, val.csv, test.csv. The format of the csv file is:

path_to_video_1 label_1
path_to_video_2 label_2
path_to_video_3 label_3
...
path_to_video_N label_N

The original annotations can be found at the kinetics website, or you can directly use download links of kinetics-400 annotations and kinetics-600 annotations. The official annotations support two different types of files: csv and json. However, both of them don't meet the above format. Therefore, I also provide a python code to transfer json files to the corresponding csv files with correct format. It takes two inputs: the container path of all videos, the path of official json annotation files. The output annotations will be named as 'output_XXX.csv' and located at the same folder. The label-to-id mapping dictionary will be saved as 'label2id.json'. The following command is my example.

python kinetics_annotation.py --train_path /home/kaihua/datasets/kinetics-train/ \
    --test_path /home/kaihua/datasets/kinetics-test/ \
    --val_path /home/kaihua/datasets/kinetics-val/ \
    --anno_path /home/kaihua/datasets/kinetics400-anno/
Owner
Kaihua Tang
@kaihuatang.github.io/
Kaihua Tang
Flexible Networks for Learning Physical Dynamics of Deformable Objects (2021)

Flexible Networks for Learning Physical Dynamics of Deformable Objects (2021) By Jinhyung Park, Dohae Lee, In-Kwon Lee from Yonsei University (Seoul,

Jinhyung Park 0 Jan 09, 2022
Generative Exploration and Exploitation - This is an improved version of GENE.

GENE This is an improved version of GENE. In the original version, the states are generated from the decoder of VAE. We have to check whether the gere

33 Mar 23, 2022
A deep learning CNN model to identify and classify and check if a person is wearing a mask or not.

Face Mask Detection The Model is designed to check if any human is wearing a mask or not. Dataset Description The Dataset contains a total of 11,792 i

1 Mar 01, 2022
Official implementation of Deep Burst Super-Resolution

Deep-Burst-SR Official implementation of Deep Burst Super-Resolution Publication: Deep Burst Super-Resolution. Goutam Bhat, Martin Danelljan, Luc Van

Goutam Bhat 113 Dec 19, 2022
This is a project based on ConvNets used to identify whether a road is clean or dirty. We have used MobileNet as our base architecture and the weights are based on imagenet.

PROJECT TITLE: CLEAN/DIRTY ROAD DETECTION USING TRANSFER LEARNING Description: This is a project based on ConvNets used to identify whether a road is

Faizal Karim 3 Nov 06, 2022
Selene is a Python library and command line interface for training deep neural networks from biological sequence data such as genomes.

Selene is a Python library and command line interface for training deep neural networks from biological sequence data such as genomes.

Troyanskaya Laboratory 323 Jan 01, 2023
Self-Supervised Learning with Kernel Dependence Maximization

Self-Supervised Learning with Kernel Dependence Maximization This is the code for SSL-HSIC, a self-supervised learning loss proposed in the paper Self

DeepMind 29 Dec 29, 2022
Using contrastive learning and OpenAI's CLIP to find good embeddings for images with lossy transformations

The official code for the paper "Inverse Problems Leveraging Pre-trained Contrastive Representations" (to appear in NeurIPS 2021).

Sriram Ravula 26 Dec 10, 2022
[NeurIPS-2020] Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID.

Self-paced Contrastive Learning (SpCL) The official repository for Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID

Yixiao Ge 286 Dec 21, 2022
Compressed Video Action Recognition

Compressed Video Action Recognition Chao-Yuan Wu, Manzil Zaheer, Hexiang Hu, R. Manmatha, Alexander J. Smola, Philipp Krähenbühl. In CVPR, 2018. [Proj

Chao-Yuan Wu 479 Dec 26, 2022
Algorithms for outlier, adversarial and drift detection

Alibi Detect is an open source Python library focused on outlier, adversarial and drift detection. The package aims to cover both online and offline d

Seldon 1.6k Dec 31, 2022
Offcial repository for the IEEE ICRA 2021 paper Auto-Tuned Sim-to-Real Transfer.

Offcial repository for the IEEE ICRA 2021 paper Auto-Tuned Sim-to-Real Transfer.

47 Jun 30, 2022
Dynamic Graph Event Detection

DyGED Dynamic Graph Event Detection Get Started pip install -r requirements.txt TODO Paper link to arxiv, and how to cite. Twitter Weather dataset tra

Mert Koşan 3 May 09, 2022
The official repository for BaMBNet

BaMBNet-Pytorch Paper

Junjun Jiang 18 Dec 04, 2022
FEDn is an open-source, modular and ML-framework agnostic framework for Federated Machine Learning

FEDn is an open-source, modular and ML-framework agnostic framework for Federated Machine Learning (FedML) developed and maintained by Scaleout Systems. FEDn enables highly scalable cross-silo and cr

Scaleout 75 Nov 09, 2022
Light-weight network, depth estimation, knowledge distillation, real-time depth estimation, auxiliary data.

light-weight-depth-estimation Boosting Light-Weight Depth Estimation Via Knowledge Distillation, https://arxiv.org/abs/2105.06143 Junjie Hu, Chenyou F

Junjie Hu 13 Dec 10, 2022
ICON: Implicit Clothed humans Obtained from Normals (CVPR 2022)

ICON: Implicit Clothed humans Obtained from Normals Yuliang Xiu · Jinlong Yang · Dimitrios Tzionas · Michael J. Black CVPR 2022 News 🚩 [2022/04/26] H

Yuliang Xiu 1.1k Jan 04, 2023
Harmonious Textual Layout Generation over Natural Images via Deep Aesthetics Learning

Harmonious Textual Layout Generation over Natural Images via Deep Aesthetics Learning Code for the paper Harmonious Textual Layout Generation over Nat

7 Aug 09, 2022
Official repository for the ICCV 2021 paper: UltraPose: Synthesizing Dense Pose with 1 Billion Points by Human-body Decoupling 3D Model.

UltraPose: Synthesizing Dense Pose with 1 Billion Points by Human-body Decoupling 3D Model Official repository for the ICCV 2021 paper: UltraPose: Syn

MomoAILab 92 Dec 21, 2022
code for Grapadora research paper experimentation

Road feature embedding selection method Code for research paper experimentation Abstract Traffic forecasting models rely on data that needs to be sens

Eric López Manibardo 0 May 26, 2022