Dr. Ajay Divakaran Profile

Dr. Ajay Divakaran

Group Manager at SRI International

SPIE Involvement:

Author

Publications (32)

Proceedings Article | 7 May 2010 Paper

Wide area active collaborative tracking of waterborne vessels

Amir Tamrakar, Sang-Hack Jung, Christopher Broaddus, Ajay Divakaran, Harpreet Sawhney

Proceedings Volume 7666, 76662H (2010) https://doi.org/10.1117/12.850467

KEYWORDS: Cameras, Sensors, Imaging systems, Scanning tunneling microscopy, Target detection, Automatic tracking, Detection and tracking algorithms, Computer programming, Visibility, Databases

Read Abstract +

Proceedings Article | 5 May 2010 Paper

Weapon identification across varying acoustic conditions using an exemplar embedding approach

Saad Khan, Ajay Divakaran, Harpreet Sawhney

Proceedings Volume 7666, 76662B (2010) https://doi.org/10.1117/12.850185

KEYWORDS: Weapons, Acoustics, Firearms, Distance measurement, Feature extraction, Bridges, Databases, Pollution control, Principal component analysis, Data modeling

Read Abstract +

Gunshot recordings have the potential for both tactical detection and forensic evaluation particularly to ascertain information about the type of firearm and ammunition used. Perhaps the most significant challenge to such an analysis is the effect of recording conditions on the audio signature of recorded data. In this paper we present a first study of using an exemplar embedding approach to automatically detect and classify firearm type across different recording conditions. We demonstrate that a small number of exemplars can span the space of gunshot audio signatures and that this optimal set can be obtained using a wrapper function. By projecting a given gunshot to the subspace spanned by the exemplar set a distance measure/feature vector is obtained that enables comparisons across recording conditions. We also investigate the use of a hierarchy of gunshot classifications that assists in improving finer level classification by pruning out gunshot labeling that is inconsistent with its higher level type. The embedding based approach can thus be used both by itself and as a pruning stage for other search techniques. Our dataset includes 20 different gun types captured in a number of different conditions. This data acts as our original exemplar set. The dataset also includes 12 gun types each with multiple shots recorded in the same conditions as the exemplar set. This second set provides our training and testing sets. We show that we can reduce our exemplar space from 20 to only 4 uniquely different gunshots without significantly limiting the ability of our embedding approach to discriminate different gunshots in the training and testing sets. The basic hypothesis in the embedding approach is that the relationship between the set of exemplars and space of gunshots including the testing/training set would be robust to a change in recording conditions or the environment. That is to say the embedding distance between a particular gunshot and the exemplars would tend to remain the same in changing environments. The implication of this are two-fold; first, unlike other dimensionality reduction approaches we have access to particular instances/examples of entities (the exemplars), which act as bridges to connect different recording conditions. Second, the embedding distances are invariant across recording conditions, the embedded vector can be used as a feature of similarity between gunshots recorded in different conditions. Unlike other dimensionality reduction approaches , our approach generates descriptions that are always in terms of the same exemplars. In other approaches such as PCA, the data driven nature makes it difficult if not impossible to make correspondence in the dimensions in one space to another. We have shown that gunshot classification across different recording conditions can be performed at a reasonable degree of certainty (60-72%) at a finer level (gunshot to weapon model) and at a high degree of certainty (95-100%) at a higher degree of abstraction (gunshot to `handgun' or `rifle'). We also investigate the use of simulated recording conditions and artificial noise to quantitatively evaluate the performance of our approach.

Proceedings Article | 15 April 2010 Paper

Long range audio and audio-visual event detection using a laser Doppler vibrometer

Tao Wang, Zhigang Zhu, Ajay Divakaran

Proceedings Volume 7704, 77040J (2010) https://doi.org/10.1117/12.853593

KEYWORDS: Laser Doppler velocimetry, Video, Cameras, Signal detection, Target detection, Acoustics, Calibration, Sensors, Doppler effect, Metals

Read Abstract +

Proceedings Article | 5 May 2009 Paper

Weapon identification using hierarchical classification of acoustic signatures

Saad Khan, Ajay Divakaran, Harpreet Sawhney

Proceedings Volume 7305, 730510 (2009) https://doi.org/10.1117/12.818375

KEYWORDS: Weapons, Firearms, Feature extraction, Acoustics, Databases, Video, Data modeling, Homeland security, Solids, Machine learning

Read Abstract +

Proceedings Article | 13 April 2009 Paper

ACT-Vision: active collaborative tracking for multiple PTZ cameras

Christopher Broaddus, Thomas Germano, Nicholas Vandervalk, Ajay Divakaran, Shunguang Wu, Harpreet Sawhney

Proceedings Volume 7345, 734505 (2009) https://doi.org/10.1117/12.818378

KEYWORDS: Cameras, Sensors, Imaging systems, Global Positioning System, Video, Target detection, Detection and tracking algorithms, Visibility, 3D modeling, Radar

Read Abstract +

Proceedings Article | 19 January 2009 Paper

Discriminative genre-independent audio-visual scene change detection

Kevin Wilson, Ajay Divakaran

Proceedings Volume 7255, 725502 (2009) https://doi.org/10.1117/12.805624

KEYWORDS: Video, Visualization, Multimedia, Data modeling, Mahalanobis distance, Semantic video, Sensors, Feature extraction, Systems modeling, Binary data

Read Abstract +

Proceedings Article | 28 January 2008 Paper

Audio scene segmentation for video with generic content

Feng Niu, Naveen Goela, Ajay Divakaran, Mohamed Abdel-Mottaleb

Proceedings Volume 6820, 68200S (2008) https://doi.org/10.1117/12.760267

KEYWORDS: Video, Video surveillance, Data modeling, Analytical research, Information visualization, Image segmentation, Systems modeling, FDA class I medical device development, Visualization, Semantic video

Read Abstract +

Proceedings Article | 29 January 2007 Paper

Assessment of end-user response to sports highlights extraction for personal video recorders

Hal Shubin, Ajay Divakaran, Kent Wittenburg, Kadir Peker, Regunathan Radhakrishnan

Proceedings Volume 6506, 650605 (2007) https://doi.org/10.1117/12.713381

KEYWORDS: Video, Human-machine interfaces, Multimedia, Detection and tracking algorithms, Imaging systems, Electronic imaging, Digital recording

Read Abstract +

Proceedings Article | 16 January 2006 Paper

Blind summarization: content-adaptive video summarization using time-series analysis

Ajay Divakaran, Regunathan Radhakrishnan, Kadir Peker

Proceedings Volume 6073, 607302 (2006) https://doi.org/10.1117/12.648419

KEYWORDS: Video, Video surveillance, Surveillance, Feature extraction, Sensors, Analytical research, Statistical analysis, Consumer electronics, Multimedia, Video processing

Read Abstract +

Severe complexity constraints on consumer electronic devices motivate us to investigate general-purpose video summarization techniques that are able to apply a common hardware setup to multiple content genres. On the other hand, we know that high quality summaries can only be produced with domain-specific processing. In this paper, we present a time-series analysis based video summarization technique that provides a general core to which we are able to add small content-specific extensions for each genre. The proposed time-series analysis technique consists of unsupervised clustering of samples taken through sliding windows from the time series of features obtained from the content. We classify content into two broad categories, scripted content such as news and drama, and unscripted content such as sports and surveillance. The summarization problem then reduces to finding either finding semantic boundaries of the scripted content or detecting highlights in the unscripted content. The proposed technique is essentially an event detection technique and is thus best suited to unscripted content, however, we also find applications to scripted content. We thoroughly examine the trade-off between content-neutral and content-specific processing for effective summarization for a number of genres, and find that our core technique enables us to minimize the complexity of the content-specific processing and to postpone it to the final stage. We achieve the best results with unscripted content such as sports and surveillance video in terms of quality of summaries and minimizing content-specific processing. For other genres such as drama, we find that more content-specific processing is required. We also find that judicious choice of key audio-visual object detectors enables us to minimize the complexity of the content-specific processing while maintaining its applicability to a broad range of genres. We will present a demonstration of our proposed technique at the conference.

Proceedings Article | 16 January 2006 Paper

Subjective assessment of consumer video summarization

Clifton Forlines, Kadir Peker, Ajay Divakaran

Proceedings Volume 6073, 60730J (2006) https://doi.org/10.1117/12.648554

KEYWORDS: Video, Televisions, Human-machine interfaces, Analytical research, Video surveillance, Multimedia, Document management, Consumer electronics, Control systems, Digital video discs

Read Abstract +

Proceedings Article | 25 October 2005 Paper

Browsing news and talk video on a consumer electronics platform using face detection

Kadir Peker, Ajay Divakaran, Tom Lanning

Proceedings Volume 6015, 601519 (2005) https://doi.org/10.1117/12.630953

KEYWORDS: Video, Facial recognition systems, Consumer electronics, Sensors, Cameras, Human-machine interfaces, Visualization, Semantic video, Image filtering, Digital video recorders

Read Abstract +

Proceedings Article | 14 March 2005 Paper

Systematic acquisition of audio classes for elevator surveillance

Regunathan Radhakrishnan, Ajay Divakaran

Proceedings Volume 5685, (2005) https://doi.org/10.1117/12.587814

KEYWORDS: Surveillance, Time series analysis, Video, Video surveillance, Feature extraction, Sensors, Data acquisition, Classification systems, Statistical analysis, Data modeling

Read Abstract +

Proceedings Article | 17 January 2005 Paper

Modeling sports highlights using a time-series clustering framework and model interpretation

Regunathan Radhakrishnan, Isao Otsuka, Ziyou Xiong, Ajay Divakaran

Proceedings Volume 5682, (2005) https://doi.org/10.1117/12.588059

KEYWORDS: Data modeling, Mining, Modeling, Statistical analysis, Video, Multimedia, Statistical modeling, Target detection, Feature extraction, Matrices

Read Abstract +

Proceedings Article | 18 January 2004 Paper

MPEG-7 meta-data enhanced encoder system for embedded systems

Kohtaro Asai, Hirofumi Nishikawa, Daiki Kudo, Ajay Divakaran

Proceedings Volume 5308, (2004) https://doi.org/10.1117/12.538253

KEYWORDS: Video, Digital video discs, Computer programming, Feature extraction, Embedded systems, Video compression, Detection and tracking algorithms, Human-machine interfaces, Information technology, Analytical research

Read Abstract +

Proceedings Article | 18 December 2003 Paper

Audio-visual event detection based on mining of semantic audio-visual labels

King-Shy Goh, Koji Miyahara, Regunathan Radhakrishnan, Ziyou Xiong, Ajay Divakaran

Proceedings Volume 5307, (2003) https://doi.org/10.1117/12.524572

KEYWORDS: Video, Visualization, Curium, Databases, Televisions, Feature extraction, Mining, Video surveillance, Semantic video, Signal detection

Read Abstract +

Proceedings Article | 18 December 2003 Paper

Video mining using combinations of unsupervised and supervised learning techniques

Ajay Divakaran, Koji Miyahara, Kadir Peker, Regunathan Radhakrishnan, Ziyou Xiong

Proceedings Volume 5307, (2003) https://doi.org/10.1117/12.532786

KEYWORDS: Video, Video surveillance, Mining, Machine learning, Surveillance, Feature extraction, Digital video discs, Video processing, Data mining, Binary data

Read Abstract +

Proceedings Article | 26 November 2003 Paper

An extended framework for adaptive playback-based video summarization

Kadir Peker, Ajay Divakaran

Proceedings Volume 5242, (2003) https://doi.org/10.1117/12.514742

KEYWORDS: Video, Visualization, Cameras, Facial recognition systems, Video surveillance, Motion measurement, Visual system, Motion detection, Sensors, Image segmentation

Read Abstract +

Proceedings Article | 26 November 2003 Paper

Investigation on effectiveness of mid-level feature representation for semantic boundary detection in news video

Regunathan Radhakrishan, Ziyou Xiong, Ajay Divakaran, Bhiksha Raj

Proceedings Volume 5242, (2003) https://doi.org/10.1117/12.514397

KEYWORDS: Video, Semantic video, Feature extraction, Optical tracking, Signal detection, Digital filtering, Speech recognition, Visualization, Machine learning, Systems modeling

Read Abstract +

Proceedings Article | 10 January 2003 Paper

Automatic extraction of soccer video highlights using a combination of motion and audio features

Romain Cabasson, Ajay Divakaran

Proceedings Volume 5021, (2003) https://doi.org/10.1117/12.476291

KEYWORDS: Video, Feature extraction, Video compression, Cameras, Motion measurement, Binary data, Digital filtering, Computing systems, Motion estimation, Image segmentation

Read Abstract +

Proceedings Article | 10 January 2003 Paper

Procedure for audio-assisted browsing of news video using generalized sound recognition

Ajay Divakaran, Regunathan Radhakrishnan, Ziyou Xiong, Michael Casey

Proceedings Volume 5021, (2003) https://doi.org/10.1117/12.476294

KEYWORDS: Video, Feature extraction, Video compression, Optical tracking, Semantic video, Visualization, Databases, Computing systems, Digital filtering, Motion analysis

Read Abstract +

Proceedings Article | 10 December 2002 Paper

Video browsing system for personal video recorders

Ajay Divakaran, Kohtaro Asai

Proceedings Volume 4861, (2002) https://doi.org/10.1117/12.470201

KEYWORDS: Video, Visualization, Human-machine interfaces, Video compression, Semantic video, Information technology, Computer programming, Visual compression, Multimedia, Feature extraction

Read Abstract +

Proceedings Article | 1 July 2002 Paper

Framework for measurement of the intensity of motion activity of video segments

Kadir Peker, Ajay Divakaran

Proceedings Volume 4862, (2002) https://doi.org/10.1117/12.473029

KEYWORDS: Video, Motion measurement, Error analysis, Video surveillance, Quantization, Cameras, Motion analysis, Image segmentation, Motion estimation, Statistical analysis

Read Abstract +

Proceedings Article | 19 December 2001 Paper

Rapid generation of sports video highlights using the MPEG-7 motion activity descriptor

Kadir Peker, Romain Cabasson, Ajay Divakaran

Proceedings Volume 4676, (2001) https://doi.org/10.1117/12.451102

KEYWORDS: Video, Video compression, Motion measurement, Cameras, Visualization, Feature extraction, Head, Databases, Lithium, Digital filtering

Read Abstract +

Proceedings Article | 19 December 2001 Paper

Data reduction procedure for principal cast and other talking head detection

Ajay Divakaran, Regunathan Radhakrishnan

Proceedings Volume 4676, (2001) https://doi.org/10.1117/12.451089

KEYWORDS: Head, Video, Facial recognition systems, Data modeling, Video compression, Feature extraction, Distance measurement, Video processing, Motion detection, Raster graphics

Read Abstract +

Proceedings Article | 12 November 2001 Paper

MPEG-7 applications for video browsing and analysis

Ajay Divakaran, Miroslaw Bober, Kohtaro Asai

Proceedings Volume 4518, (2001) https://doi.org/10.1117/12.448223

KEYWORDS: Video, Shape analysis, Multimedia, Cameras, Computing systems, Standards development, Object recognition, Information technology, Motion analysis, Motion detection

Read Abstract +

SPIE Journal Paper | 1 October 2001

Video summarization using descriptors of motion activity: a motion activity based approach to key-frame extraction from video shots

Ajay Divakaran, Regunathan Radhakrishnan, Kadir Peker

JEI, Vol. 10, Issue 04, (October 2001) https://doi.org/10.1117/12.10.1117/1.1406507

KEYWORDS: Video, Motion measurement, Visualization, Video surveillance, Nonuniform sampling, Feature extraction, Head, Video compression, Distance measurement, Cameras

Proceedings Article | 22 March 2001 Paper

MPEG-4/7-based Internet video and still-image browsing system

Miroslaw Bober, Kohtaro Asai, Ajay Divakaran

Proceedings Volume 4209, (2001) https://doi.org/10.1117/12.420843

KEYWORDS: Video, Internet, Feature extraction, Multimedia, Shape analysis, Standards development, Information technology, Video compression, Java, Image transmission

Read Abstract +

Proceedings Article | 1 January 2001 Paper

Video summarization using motion descriptors

Ajay Divakaran, Kadir Peker, Huifang Sun

Proceedings Volume 4315, (2001) https://doi.org/10.1117/12.410963

KEYWORDS: Video, Video surveillance, Video compression, Optical filters, Digital filtering, Feature extraction, Data processing, Image segmentation, Sun, Databases

Read Abstract +

Proceedings Article | 1 January 2001 Paper

Automatic measurement of intensity of motion activity of video segments

Kadir Peker, Ajay Divakaran, Thomas Papathomas

Proceedings Volume 4315, (2001) https://doi.org/10.1117/12.410944

KEYWORDS: Video, Video surveillance, Motion measurement, Quantization, Cameras, Error analysis, Motion analysis, Video compression, Semantic video, Analytical research

Read Abstract +

Proceedings Article | 23 December 1999 Paper

Descriptor for spatial distribution of motion activity for compressed video

Ajay Divakaran, Huifang Sun

Proceedings Volume 3972, (1999) https://doi.org/10.1117/12.373571

KEYWORDS: Video, Video surveillance, Motion measurement, Computer programming, Feature extraction, Video compression, Semantic video, Surveillance, Databases, Head

Read Abstract +

Proceedings Article | 23 December 1999 Paper

Fade-in/out scene change detection in the MPEG-1/2/4 compressed video domain

Ajay Divakaran, Hiroshi Ito, Huifang Sun, Tommy Poon

Proceedings Volume 3972, (1999) https://doi.org/10.1117/12.373584

KEYWORDS: Video compression, Video, Computer programming, Video processing, Image segmentation, Detection and tracking algorithms, Sun, Information technology, Televisions, Image compression

Read Abstract +

Proceedings Article | 17 December 1998 Paper

Scene change detection and feature extraction for MPEG-4 sequences

Ajay Divakaran, Hiroshi Ito, Huifang Sun, Tommy Poon

Proceedings Volume 3656, (1998) https://doi.org/10.1117/12.333874

KEYWORDS: Video, Video compression, Feature extraction, Motion measurement, Image retrieval, Image segmentation, Video processing, Image compression, Image storage, Databases

Read Abstract +