PKU-Spike-A Dataset

The PKU-Spike-A dataset is constructed by National Engineering Laboratory for Video Technology (NELVT), Peking University, sponsored by the National Basic Research Program of China and Chinese National Natural Science Foundation. The goals to create the PKU-Spike-A include:
(1) providing the worldwide researchers of the spike coding community a large-scale spike dataset for evaluating their coding algorithms;
(2) facilitating the development of high-efficient spike coding technologies by providing large-scale spike sequences with different depth of field or at different luminance and time conditions.
Therefore, the PKU-Spike-A dataset is now partly made available for the academic purpose only on a case-by-case basis. The first available part contains Class A (normal speed) and Class B (high speed) with six spike sequences in total. More sequences are under evaluation which will be public available in the near future.

Sequence Time-length (s) Total spike number
Class A
(normal speed)
office 3.84 163679049
rolling 3.84 464711432
wavehand 3.84 845044270
Class B
(high speed)
fork 3.84 125627228
disk-pku 3.84 535852602
rotation 3.84 930631025

The PKU-Spike-A dataset is constructed by National Engineering Laboratory for Video Technology (NELVT), Peking University, sponsored by the National Basic Research Program of China and Chinese National Natural Science Foundation. The NELVT at Peking University is serving as the technical agent for distribution of the dataset and reserves the copyright of all the sequences in the dataset. Any researcher who requests the PKU-Spike-A dataset must sign this agreement and thereby agrees to observe the restrictions listed in this document. Failure to observe the restrictions will result in access being denied for the request of the future version of the PKU-Spike-A dataset and being subject to civil damages in the case of publication of sequences that have not been approved for release.

LICENSE

  • The spike sequences for download are part of the PKU-Spike-A dataset.
  • The sequences can only be used for ACADEMIC PURPOSES. NO COMERCIAL USE is allowed.
  • Copyright © National Engineering Laboratory for Video Technology (NELVT) and Institute of Digital Media, Peking University (PKU-IDM). All rights reserved.

Download

  • You can download the agreement(pdf) by clicking the DOWNLOAD link (available soon).
  • After filling it, please send the electrical version to our Email: pkuml at pku.edu.cn (Subject: PKU-Spike-A Agreement) and mail the paper version to our lab: Room 2604, Science Building No. 2, Peking University, No.5 Yiheyuan Road, Haidian District, Beijing, P.R.China .

PKU-DVS Dataset

The PKU-DVS dataset is constructed by National Engineering Laboratory for Video Technology (NELVT), Peking University, sponsored by the National Basic Research Program of China and Chinese National Natural Science Foundation. The goals to create the PKU-DVS include:
(1) providing the worldwide researchers of the spike coding community a large-scale DVS spike dataset for evaluating their coding algorithms;
(2) facilitating the development of high-efficient spike coding technologies by providing large-scale spike sequences with different depth of field or at different luminance and time conditions.
Therefore, the PKU-DVS dataset is now partly made available for the academic purpose only on a case-by-case basis. The first available part contains Class A (indoor) and Class B (outdoor) with seven and six spike sequences, respectively. More sequences are under evaluation which will be public available in the near future.

Sequence Time (s) Total spike number
Class A: indoor waterdrop 3.80 11563244
fluorescent 5.44 11643175
lighter 2.10 2792140
football 7.21 9745102
jump 3.28 2375023
game 9.57 5918278
pendulum 5.37 113683
Class B: outdoor intersection 10.35 30483325
pedestrians 355.21 25455049
daytime-traffic1 109.84 14246080
daytime-traffic2 301.60 5525454
night-roadside 63.37 17018998
night-traffic 15.47 5423636

The PKU-DVS dataset is constructed by National Engineering Laboratory for Video Technology (NELVT), Peking University, sponsored by the National Basic Research Program of China and Chinese National Natural Science Foundation. The NELVT at Peking University is serving as the technical agent for distribution of the dataset and reserves the copyright of all the sequences in the dataset. Any researcher who requests the PKU-DVS dataset must sign this agreement and thereby agrees to observe the restrictions listed in this document. Failure to observe the restrictions will result in access being denied for the request of the future version of the PKU-DVS dataset and being subject to civil damages in the case of publication of sequences that have not been approved for release.

LICENSE

  • The spike sequences for download are part of the PKU-DVS dataset.
  • The sequences can only be used for ACADEMIC PURPOSES. NO COMERCIAL USE is allowed.
  • Copyright © National Engineering Laboratory for Video Technology (NELVT) and Institute of Digital Media, Peking University (PKU-IDM). All rights reserved.

Download

  • You can download the agreement(pdf) by clicking the DOWNLOAD link (available soon).
  • After filling it, please send the electrical version to our Email: pkuml at pku.edu.cn (Subject: PKU-DVS Agreement)

PKU HumanID Dataset

The PKU HumanDI dataset is constructed by National Engineering Laboratory for Video Technology (NELVT), Peking University, sponsored by the National Basic Research Program of China and Chinese National Natural Science Foundation.

This dataset is composed of videos subjects crossing 11 cameras in a campus. It includes 6 high definition network cameras (Camera HD01, Camera HD02, Camera HD03, Camera HD04, Camera HD05, Camera HD06) and 7 normal network cameras (Camera BWBQ, Camera DCM, Camera WMHD, Camera XDMN, Camera YGLN, Camera YGLQ, Camera YTX). Some samples of the labeled results are shown below:

pku-humanid-dataset-1

HD Cameras (HD 01, HD 02, HD 04, HD 06)

pku-humanid-dataset-2

Normal Cameras (WMHD, DCM, YTX, YGLN)

The PKU HumanID dataset is now partly made available for the academic purpose only on a case-by-case basis. The NELVT at Peking University is serving as the technical agent for distribution of the dataset and reserves the copyright of all the videos in the dataset. Any researcher who requests the PKU HumanID dataset must sign this agreement and thereby agrees to observe the restrictions listed in this document.

LICENSE

  • The videos and the corresponding annotation results for download are part of PKU HumanID.
  • The videos and the corresponding annotation results can only be used for ACADEMIC PURPOSES. NO COMERCIAL USE is allowed.
  • Copyright © National Engineering Laboratory for Video Technology (NELVT) and Institute of Digital Media, Peking University (PKU-IDM). All rights reserved.

All publications using PKU HumanID dataset should cite the paper below:

  • Lan Wei, Yonghong Tian, Yaowei Wang, Tiejun Huan, Swiss-System Based Cascade Ranking for Gait-based Person Re-identification, 29th AAAI Conference on Artificial Intelligence, Austin Texas, USA, January, 2015

DOWNLOAD

You can download the agreement (pdf) by clicking the DOWNLOAD link.

After filling it, please send the electrical version to our Email: pkuml at pku.edu.cn (Subject: PKU-HumanID-Agreement)

After confirming your information, we will send the download link and password to you via Email. You need to follow the agreement.

Image saliency: From intrinsic to extrinsic context

An implementation of “Wang M, Konrad J, Ishwar P, Jing K, Rowley H (2011) Image saliency: From intrinsic to extrinsic context. CVPR, 2011.”

Re-implemented by Jia Li (jiali@buaa.edu.cn) and Shu Fang (sfang@pku.edu.cn).

Code folder: contains our implementation of (Wang et al. 2011) and the metrics for computing AUC, EOF and FS. More details can be found in our paper submitted to IJCV (J. Li et al. Measuring Visual Surprise Jointly from Intrinsic and Extrinsic Contexts for Image Saliency Estimation)

MIT1003 folder: the data used for testing (Wang et al. 2011). Subfolder “image” contains all the images from the dataset MIT1003, and subfolder “refImages” contains all the 20 most similar images retrieved from a large database with 31.2 million images.

Result folder: three subfolders, IES_intSal, IES_extSal and IES, contain the saliency maps from intrinsic context, extrinsic context and both contexts, respectively.

 

Code:
/mlg/download/code/wang11.zip
/mlg/download/code/wang11-codeResult.zip

PKU-RSD Dataset

We constructed this PKU-RSD (Regional Saliency Dataset) dataset that could capture spatiotemporal visual saliency for evaluating different video saliency models. This dataset contains 431 short videos, which cover various scenes (surveillance, ad, news, cartoon, movie etc.) and the corresponding annotation results of salient objects in sampled key frames manually labeled by 23 subjects. Some samples of the annotation results are shown below:

samples of RSD

LICENSE

  • The videos and the corresponding annotation results for download are part of PKU-RSD (Regional Saliency Dataset) dataset.
  • The videos and the corresponding annotation results can only be used for ACADEMIC PURPOSES. NO COMERCIAL USE is allowed.
  • Copyright © National Engineering Laboratory for Video Technology (NELVT) and Institute of Digital Media, Peking University (PKU-IDM). All rights reserved.

All publications using PKU-RSD should cite the paper below:

  • Jia Li, Yonghong Tian, Tiejun Huang, Wen Gao. A DATASET AND EVALUATION METHODOLOGY FOR VISUAL SALIENCY IN VIDEO. ICME 2009

DOWNLOAD
You can download the agreement (pdf) by clicking the DOWNLOAD link.
After filling it, please send the electrical version to our Email: pkuml at pku.edu.cn (Subject: PKU-RSD Agreement)
After confirming your information, we will send the download link and password to you via Email. You need to follow the agreement.

北京大学数字媒体所媒体学习组招生简章

招生信息

[New]课题组持续招收具有计算神经科学、数学和机器学习背景的博士生、博士后和研究人员,有意愿者请联系!

[Hot]课题组持续从北京地区高校中招收大二、大三的Intern,有愿意从事分布式机器学习、神经形态视觉和视频大数据等领域相关研究的同学(特别是有意愿攻读博士学位的,数学、英语、编程与创新能力强者),请与我们联系!

博士(含直博)招生方向及导师

招生专业
研究方向
导师
招生简章对应科目
招生单位
计算机应用技术
分布式机器学习、神经形态视觉和视频大数据
田永鸿
北大计算机学院:
视觉信息处理与类脑智能
脉冲神经网络
北大计算机学院

注:“招生简章对应科目”是指对应于招生简章中的科目(黑体为首选科目),需要考生在报考时填写的名称。考生在报考是应同时填写希望报考的导师。

学术硕士招生方向及导师

招生专业
研究方向
导师
招生简章对应科目
招生单位
计算机应用技术
机器学习与媒体大数据分析 田永鸿 02. 多媒体信息处理技术
北大深研院信息工程学院

注:“招生简章对应科目”是指对应于招生简章中的科目(黑体为首选科目),需要考生在报考时填写的名称。考生在报考时应同时填写希望报考的导师。

联系方式
Email: yhtian (at) pku.edu.cn
Tel: +86-10-62755965

——————————————————–

PKU-SVD-A DATASET

Here is several surveillance videos of the PKU-SVD-A Dataset for download. The archive consists of six SD576 and two 1200p sequences.

Resolution Video Name Frames Frame Rate
SD576(720×576) Bank 3000 30fps
Campus
Classover
Crossroad
Office
Overbridge
1200p(1600×1200) Intersection 1000 30fps
Mainroad

Bank Bank

Bank Campus

Bank Classover

Bank Crossroad

Bank Office

Bank Overbridge

Bank Intersection

Bank Mainroad

PKU-SVD-A

To evaluate the surveillance video coding performance, we constructed a large-scale dataset, called PKU-SVD-A, by collecting 73 videos with different resolutions (ranging from SD, 720p, 1600*1200, and 1920*1080) or at different weather and time conditions (e.g., dark, fog, rain,…). This dataset will be online publically available soon for the research usage.

LICENSE

  • The videos for download is part of PKU-SVD-A (Peking University Surveillance Video Dataset A).
  • The videos can only be used for ACADEMIC PURPOSES. NO COMERCIAL USE is allowed.
  • Copyright © National Engineering Laboratory for Video Technology (NELVT) and Institute of Digital Media, Peking University (PKU-IDM). All rights reserved.

All publications using PKU-SVD-A should cite the papers below:

  1. X. Zhang, Y. Tian, T. Huang, S. Dong, W. Gao, Optimizing the hierarchical prediction and coding in HEVC for surveillance and conference videos with background modeling, IEEE Trans. on Image Processing, 2014.
  2. X. Zhang, T. Huang, Y. Tian, W. Gao, Background Modeling Based Adaptive Prediction for Surveillance Video Coding, IEEE Trans. on Image Processing, 2014.
  3. W. Gao, Y. Tian, T. Huang, S. Ma, X. Zhang, IEEE 1857 Standard Empowering Smart Video Surveillance Systems, IEEE Intelligent Systems, 2013

Download

  • You can download the agreement(pdf) by clicking the DOWNLOAD link.
  • After filling it, please send the electrical version to our Email: pkuml at pku.edu.cn (Subject: PKU-SVD-A Agreement) .Please send it through an academic or institute email-addresses such as xxx at xxx.edu.xx. Requests from free email addresses (outlook, gmail, qq etc) will be kindly refused.

PKU-EAQA DATASET

To compare the performances of different enhancement algorithms, we constructed this PKU-EAQA (Peking University Enhancement Algorithm Quality Assessment) dataset. This dataset contains 300 images in bad visibility (in haze, low light, etc.), 1500 enhanced images generated by different enhancement algorithms and their subjective quality assessment results.

LICENSE

  • The images and the subjective quality assessment results for download are part of PKU-EAQA (Peking University Enhancement Algorithm Quality Assessment) dataset.
  • The images and the subjective quality assessment results can only be used for ACADEMIC PURPOSES. NO COMERCIAL USE is allowed.
  • Copyright © National Engineering Laboratory for Video Technology (NELVT) and Institute of Digital Media, Peking University (PKU-IDM). All rights reserved.

All publications using PKU-EAQA should cite the papers below:

  • Zhengying Chen, Tingting Jiang, Yonghong Tian, Quality Assessment for Comparing Image Enhancement Algorithms. To appear in IEEE Conference on Computer Vision and Pattern Recognition, 2014.

CVPR2014-Poster-PaperID1469
DOWNLOAD
You can download the agreement(pdf) by clicking the DOWNLOAD link.
After filling it, please send the electrical version to our Email: pkuml at pku.edu.cn (Subject: PKU-EAQA Agreement) After confirming your information, we will send the download link and password to you via Email. You need to follow the agreement.

PKU-VISUAL-OBJECTS DATASET

The PKU-VISUAL-OBJECTS dataset is constructed by National Engineering Laboratory for Video Technology (NELVT), Peking University, sponsored by the National Basic Research Program of China and Chinese National Natural Science Foundation.
the PKU-VISUAL-OBJECTS dataset is now partly made available for the academic purpose only on a case-by-case basis. The NELVT at Peking University is serving as the technical agent for distribution of the dataset and reserves the copyright of all the videos in the dataset. Any researcher who requests the PKU-VISUAL-OBJECTS dataset must sign this agreement and thereby agrees to observe the restrictions listed in this document. Failure to observe the restrictions will result in access being denied for the request of the future version of the PKU-VISUAL-OBJECTS dataset and being subject to civil damages in the case of publication of videos that have not been approved for release.

VIDEO MASKS

Resolution Sequence Frames Mask
1080p Kimono 238  kimono_1920x1080_24_mask238
ParkScene 238  ParkScene_1920x1080_24_mask238
BasketballDrive 499  BasketballDrive_1920x1080_50_mask499
720p Johnny 598  Johnny_1280x720_60_mask598
KristenAndSara 598  KristenAndSara_1280x720_60_mask598
FourPeople 598  FourPeople_1280x720_60_mask598

LICENSE

  • The videos for download is part of PKU-VISUAL-Objects Dataset.
  • The videos can only be used for ACADEMIC PURPOSES. For COMERCIAL USE, please contact us for authorization .
  • Copyright © National Engineering Laboratory for Video Technology (NELVT) and Institute of Digital Media, Peking University (PKU-IDM). All rights reserved.

All publications using PKU-VISUAL-OBJECTS Dataset should cite the paper below:

  • Tiejun Huang, Siwei Dong, Yonghong Tian, Representing Visual Objects in HEVC
    Coding Loop, IEEE Journal on Emerging and Selected Topics in Circuits and
    Systems, Volume 4, Issue 1, March 2014, 5-16. DOI:
    10.1109/JETCAS.2014.2298274.

Download

  • You can download the agreement(pdf) by clicking the DOWNLOAD link.
  • After filling it, please send the electrical version to our Email: pkuml at pku.edu.cn (Subject: PKU-VISUAL-OBJECTS Dataset Agreement)
  • The original videos (without masks) are part of HEVC/H.265 common test sequences, which can be downloaded from JCT-VC FTP (ftp://hevc@ftp.tnt.uni-hannover.de/testsequences).