Compressive Video Sensing: Algorithms, architectures, and applications
Abstract
The design of conventional sensors is based primarily on the Shannon?Nyquist sampling theorem, which states that a signal of bandwidth W Hz is fully determined by its discrete time samples provided the sampling rate exceeds 2 W samples per second. For discrete time signals, the Shannon?Nyquist theorem has a very simple interpretation: the number of data samples must be at least as large as the dimensionality of the signal being sampled and recovered. This important result enables signal processing in the discrete time domain without any loss of information. However, in an increasing number of applications, the Shannon-Nyquist sampling theorem dictates an unnecessary and often prohibitively high sampling rate (see "What Is the Nyquist Rate of a Video Signal?"). As a motivating example, the high resolution of the image sensor hardware in modern cameras reflects the large amount of data sensed to capture an image. A 10-megapixel camera, in effect, takes 10 million measurements of the scene. Yet, almost immediately after acquisition, redundancies in the image are exploited to compress the acquired data significantly, often at compression ratios of 100:1 for visualization and even higher for detection and classification tasks. This example suggests immense wastage in the overall design of conventional cameras.
Additional Information
© 2017 IEEE. We thank David Robert Jones for his invaluable suggestions and Doug Jones for the JAM. Richard G. Baraniuk was supported by National Science Foundation (NSF) grants CCF-1527501 and CCF-1502875, Defense Advanced Research Projects Agency (DARPA) Revolutionary Enhancement of Visibility by Exploiting Active Light-fields grant HR0011-16-C-0028, and Office of Naval Research (ONR) grant N00014-15-1-2735. Tom Goldstein was supported by NSF grant CCF-1535902 and ONR grant N00014-15-1-2676. Aswin C. Sankaranarayanan was supported by NSF grant IIS-1618823 and Army Research Office grant W911NF-16-1-0441. Christoph Studer was supported in part by Xilinx Inc. and by NSF grants ECCS-1408006 and CCF-1535897. Ashok Veeraraghavan was supported by NSF grant CCF-1527501. Michael B. Wakin was supported by NSF CAREER grant CCF-1149225 and grant CCF-1409258.Additional details
- Eprint ID
- 75846
- DOI
- 10.1109/MSP.2016.2602099
- Resolver ID
- CaltechAUTHORS:20170407-122108747
- NSF
- CCF-1527501
- NSF
- CCF-1502875
- Defense Advanced Research Projects Agency (DARPA)
- HR0011-16-C-0028
- Office of Naval Research (ONR)
- N00014-15-1-2735
- NSF
- CCF-1535902
- Office of Naval Research (ONR)
- N00014-15-1-2676
- NSF
- IIS-1618823
- Army Research Office (ARO)
- W911NF-16-1-0441
- Xilinx Inc.
- NSF
- ECCS-1408006
- NSF
- CCF-1535897
- NSF
- CCF-1527501
- NSF
- CCF-1149225
- NSF
- CCF-1409258
- Created
-
2017-04-07Created from EPrint's datestamp field
- Updated
-
2021-11-15Created from EPrint's last_modified field