Information-theoretic active scene exploration
Abstract
Studies support the need for high resolution imagery to identify persons in surveillance videos. However, the use of telephoto lenses sacrifices a wider field of view and thereby increases the uncertainty of other, possibly more interesting events in the scene. Using zoom lenses offers the possibility of enjoying the benefits of both wide field of view and high resolution, but not simultaneously. We approach this problem of balancing these finite imaging resources - or of exploration vs exploitation - using an information-theoretic approach. We argue that the camera parameters - pan, tilt and zoom - should be set to maximise information gain, or equivalently minimising conditional entropy of the scene model, comprised of multiple targets and a yet unobserved one. The information content of the former is supplied directly by the uncertainties computed using a Kalman filter tracker, while the latter is modelled using a rdquobackgroundrdquo Poisson process whose parameters are learned from extended scene observations; together these yield an entropy for the scene. We support our argument with quantitative and qualitative analyses in simulated and real-world environments, demonstrating that this approach yields sensible exploration behaviours in which the camera alternates between obtaining close-up views of the targets while paying attention to the background, especially to areas of known high activity.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In public safety or traffic video surveillance scenarios, active cameras can change their orientation and focal length by pan-tilt-zoom control to track and focus on specific targets [234], [235].
“In the public safety or traffic video surveillance scenarios, the active cameras can change its orientation and focal length by pan-tilt-zoom control to track and focus on specific targets [234], [235].”
From Empowering things with intelligence: a survey of the progress, challenges, and opportunities in artificial intelligence of things (2021) · cites this paper as [235]
Active cameras can be used in AIoT systems to capture salient and high-resolution humans and vehicles by adjusting orientation and focal length according to target location.
“Moreover, to capture salient and high-resolution humans and vehicles, active cameras can be used in AIoT systems by adjusting the orientation and focal length according to target location [234], [235].”
From Empowering things with intelligence: a survey of the progress, challenges, and opportunities in artificial intelligence of things (2021) · cites this paper as [235]
Cites
Learning patterns of activity using real-time tracking
Stauffer, C., & Grimson, W. E. L. (2000) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.868677 · Cited by 3,246 · see citing works
Bar‐Shalom, Y., Fortmann, T. E., & Cable, P. G. (1990) · The Journal of the Acoustical Society of America · doi:10.1121/1.398863 · Cited by 3,080 · see citing works
Learning the distribution of object trajectories for event recognition
Johnson, N. J., & Hogg, D. (1996) · Image and Vision Computing · doi:10.1016/0262-8856(96)01101-8 · Cited by 499 · see citing works
FRVT 2006 and ICE 2006 large-scale results
Phillips, P. J., Scruggs, W., O’Toole, A. J., et al. (2007) · doi:10.6028/nist.ir.7408 · Cited by 418 · see citing works
FRVT 2006 and ICE 2006 Large-Scale Experimental Results
Phillips, P. J., Scruggs, W., O’Toole, A. J., Flynn, P. J., Bowyer, K. W., Schott, C. L., & Sharpe, M. J. (2009) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2009.59 · Cited by 406 · see citing works
Performance evaluation of object tracking algorithms
Yin, F., Makris, D., & Velastín, S. A. (2007) · Cited by 125 · see citing works
Face cataloger: multi-scale imaging for relating identity to location
Hampapur, A., Pankanti, S., Senior, A., et al. (2004) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2003.1217896 · Cited by 98 · see citing works
Automatic learning of an activity-based semantic scene model
Makris, D., & Ellis, T. J. (2004) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2003.1217920 · Cited by 96 · see citing works
Peripheral-foveal vision for real-time object recognition and tracking in video
Gould, S. J., Arfvidsson, J., Kaehler, A., Sapp, B., Messner, M., Bradski, G. R., Baumstarck, P., Chung, S., & Ng, A. Y. (2007) · ANU Open Research (Australian National University) · Cited by 81 · see citing works
Scheduling an active camera to observe people
Costello, C. J., Diehl, C., Banerjee, A., et al. (2004) · doi:10.1145/1026799.1026808 · Cited by 77 · see citing works
Surveillance in Virtual Reality: System Design and Multi-Camera Control
Qureshi, F. Z., & Terzopoulos, D. (2007) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2007.383071 · Cited by 76 · see citing works
Information theoretic focal length selection for real-time active 3D object tracking
Denzler, Zobel, & Niemann (2003) · Proceedings Ninth IEEE International Conference on Computer Vision · doi:10.1109/iccv.2003.1238372 · Cited by 53 · see citing works
An adaptive focus-of-attention model for video surveillance and monitoring
Davis, J. W., Morison, A. M., & Woods, D. D. (2006) · Machine Vision and Applications · doi:10.1007/s00138-006-0047-x · Cited by 32 · see citing works
Acquisition of high-resolution images through on-line saccade sequence planning
Bagdanov, A. D., Bimbo, A. D., & Pernici, F. (2005) · doi:10.1145/1099396.1099419 · Cited by 26 · see citing works
A method of reactive zoom control from uncertainty in tracking
Tordoff, B. J., & Murray, D. W. (2006) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2006.09.006 · Cited by 24 · see citing works
Look there! predicting where to look for motion in an active camera network
Erdem, U. M., & Sclaroff, S. (2006) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2005.1577251 · Cited by 20 · see citing works
Dynamic action spaces for information gain maximization in search and exploration
Roy, N., & Earnest, C. M. (2006) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.1109/acc.2006.1656452 · Cited by 19 · see citing works
What Can Casual Walkers Tell Us About A 3D Scene?
Rother, D., Patwardhan, K. A., & Sapiro, G. (2007) · IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2007.4409082 · Cited by 19 · see citing works
Collaborative tracking of objects in EPTZ cameras
Bashir, F. I., & Porikli, F. (2007) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.714077 · Cited by 13 · see citing works
Hermes: A research project on human sequence evaluation
Gonzàlez, J., Roca, F. X., & Villanueva, J. J. (2026) · CRC Press eBooks · doi:10.1201/9781003759805-5 · Cited by 11 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Zhang, J., & Tao, D. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3039359 · Cited by 774 · see citing works
- Authors
- Sommerlade, E., & Reid, I.
- Affiliations
- University of Oxford (GB)
- ORCID
- Reid, Ian
- Venue
- 2008 IEEE Conference on Computer Vision and Pattern Recognition
- Publisher
- IEEE
- Published
- 2008-06
- Pages
- 1-7
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2125438290
- Cited by
- 65
- References
- 20
- Citations by year
- 2023: 1 · 2022: 1 · 2021: 1 · 2020: 3 · 2019: 1 · 2018: 1 · 2017: 5 · 2016: 5 · 2015: 4 · 2014: 8 · 2013: 10 · 2012: 4
- FWCI
- 5.61
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- Adelaide Research & Scholarship (AR&S) (University of Adelaide) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Surveillance and Tracking Methods
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
- Keywords
- information gain, conditional entropy, zoom lens, surveillance video, personal identification, exploration-exploitation balance, Kalman filter tracking, multi-target tracking
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
