15 Citations (Scopus)

Abstract

Persistent surveillance from camera networks, such as at parking lots, UAVs, etc., often results in large amounts of video data, resulting in significant challenges for inference in terms of storage, communication and computation. Compressive cameras have emerged as a potential solution to deal with the data deluge issues in such applications. However, inference tasks such as action recognition require high quality features which implies reconstructing the original video data. Much work in compressive sensing (CS) theory is geared towards solving the reconstruction problem, where state-of-the-art methods are computationally intensive and provide low-quality results at high compression rates. Thus, reconstruction-free methods for inference are much desired. In this paper, we propose reconstruction-free methods for action recognition from compressive cameras at high compression ratios of 100 and above. Recognizing actions directly from CS measurements requires features which are mostly nonlinear and thus not easily applicable. This leads us to search for such properties that are preserved in compressive measurements. To this end, we propose the use of spatiooral smashed filters, which are compressive domain versions of pixel-domain matched filters. We conduct experiments on publicly available databases and show that one can obtain recognition rates that are comparable to the oracle method in uncompressed setup, even for high compression ratios.

Original languageEnglish (US)
Article number7208864
Pages (from-to)772-784
Number of pages13
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume38
Issue number4
DOIs
StatePublished - Apr 1 2016

Fingerprint

Free Action
Imager
Image sensors
Cameras
Action Recognition
Compressive Sensing
Compression
Camera
Matched filters
Parking
Unmanned aerial vehicles (UAV)
Matched Filter
Pixels
Surveillance
Pixel
Communication
Filter
Imply
Experiments
Experiment

Keywords

  • Action recognition
  • Compressive Sensing
  • Reconstruction-free

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Software
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

Reconstruction-Free Action Inference from Compressive Imagers. / Kulkarni, Kuldeep; Turaga, Pavan.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 38, No. 4, 7208864, 01.04.2016, p. 772-784.

Research output: Contribution to journalArticle

@article{0503d75191d2484192b892a27c64c8f3,
title = "Reconstruction-Free Action Inference from Compressive Imagers",
abstract = "Persistent surveillance from camera networks, such as at parking lots, UAVs, etc., often results in large amounts of video data, resulting in significant challenges for inference in terms of storage, communication and computation. Compressive cameras have emerged as a potential solution to deal with the data deluge issues in such applications. However, inference tasks such as action recognition require high quality features which implies reconstructing the original video data. Much work in compressive sensing (CS) theory is geared towards solving the reconstruction problem, where state-of-the-art methods are computationally intensive and provide low-quality results at high compression rates. Thus, reconstruction-free methods for inference are much desired. In this paper, we propose reconstruction-free methods for action recognition from compressive cameras at high compression ratios of 100 and above. Recognizing actions directly from CS measurements requires features which are mostly nonlinear and thus not easily applicable. This leads us to search for such properties that are preserved in compressive measurements. To this end, we propose the use of spatiooral smashed filters, which are compressive domain versions of pixel-domain matched filters. We conduct experiments on publicly available databases and show that one can obtain recognition rates that are comparable to the oracle method in uncompressed setup, even for high compression ratios.",
keywords = "Action recognition, Compressive Sensing, Reconstruction-free",
author = "Kuldeep Kulkarni and Pavan Turaga",
year = "2016",
month = "4",
day = "1",
doi = "10.1109/TPAMI.2015.2469288",
language = "English (US)",
volume = "38",
pages = "772--784",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "4",

}

TY - JOUR

T1 - Reconstruction-Free Action Inference from Compressive Imagers

AU - Kulkarni, Kuldeep

AU - Turaga, Pavan

PY - 2016/4/1

Y1 - 2016/4/1

N2 - Persistent surveillance from camera networks, such as at parking lots, UAVs, etc., often results in large amounts of video data, resulting in significant challenges for inference in terms of storage, communication and computation. Compressive cameras have emerged as a potential solution to deal with the data deluge issues in such applications. However, inference tasks such as action recognition require high quality features which implies reconstructing the original video data. Much work in compressive sensing (CS) theory is geared towards solving the reconstruction problem, where state-of-the-art methods are computationally intensive and provide low-quality results at high compression rates. Thus, reconstruction-free methods for inference are much desired. In this paper, we propose reconstruction-free methods for action recognition from compressive cameras at high compression ratios of 100 and above. Recognizing actions directly from CS measurements requires features which are mostly nonlinear and thus not easily applicable. This leads us to search for such properties that are preserved in compressive measurements. To this end, we propose the use of spatiooral smashed filters, which are compressive domain versions of pixel-domain matched filters. We conduct experiments on publicly available databases and show that one can obtain recognition rates that are comparable to the oracle method in uncompressed setup, even for high compression ratios.

AB - Persistent surveillance from camera networks, such as at parking lots, UAVs, etc., often results in large amounts of video data, resulting in significant challenges for inference in terms of storage, communication and computation. Compressive cameras have emerged as a potential solution to deal with the data deluge issues in such applications. However, inference tasks such as action recognition require high quality features which implies reconstructing the original video data. Much work in compressive sensing (CS) theory is geared towards solving the reconstruction problem, where state-of-the-art methods are computationally intensive and provide low-quality results at high compression rates. Thus, reconstruction-free methods for inference are much desired. In this paper, we propose reconstruction-free methods for action recognition from compressive cameras at high compression ratios of 100 and above. Recognizing actions directly from CS measurements requires features which are mostly nonlinear and thus not easily applicable. This leads us to search for such properties that are preserved in compressive measurements. To this end, we propose the use of spatiooral smashed filters, which are compressive domain versions of pixel-domain matched filters. We conduct experiments on publicly available databases and show that one can obtain recognition rates that are comparable to the oracle method in uncompressed setup, even for high compression ratios.

KW - Action recognition

KW - Compressive Sensing

KW - Reconstruction-free

UR - http://www.scopus.com/inward/record.url?scp=84963777092&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84963777092&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2015.2469288

DO - 10.1109/TPAMI.2015.2469288

M3 - Article

VL - 38

SP - 772

EP - 784

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 4

M1 - 7208864

ER -