Sensitivity-Based Adaptive Activity Mapping for Optimal Camera Calibration

Shashank,; Indu, S.

doi:10.1007/978-981-15-0633-8_118

Cited by 1 publication

(9 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparisons of activity maps and normalized pixel sensitivity maps derived by Pan et al in [12], as [A], Mehboob et al in [13], as [B], Indu, S. in [14], as [C], and our proposed framework, as [D], are shown in Figure 5 for video dataset 1, Figure 7 for video dataset 2, Figure 9 for video dataset 3, Figure 11 for video dataset 4, Figure 13 for video dataset 5 and Figure 15 for video dataset 6, respectively. The approaches presented in [12,13] do not include any temporal relation between past and present frames.…”

Section: Discussion On Simulationsmentioning

confidence: 99%

“…However, both Pan et al in [12] and Mehboob et al in [13] failed to showcase the temporal effect of past frames on the present frame, which can be seen in Figure 5 for video dataset 1, Figure 7 for video dataset 2, Figure 9 for video dataset 3, Figure 11 for video dataset 4, Figure 13 for video dataset 5 and Figure 15 for video dataset 6, respectively. Indu, S., in [14], proposed a spatiotemporal relation between past and present frames; however, the proposed method in [14] failed to provide efficient filtering of the foreground and thus resulted in inaccurate spatiotemporal activity and sensitivity maps due to the noise and undesirable regions detected in the activity map. The proposed framework, in comparison to that of Pan et al in [12] and that of Mehboob et al in [13], presents better spatiotemporal activity tracking in terms of normalized pixel sensitivity.…”

Section: Discussion On Simulationsmentioning

confidence: 99%

“…We have compared our approach with the approaches proposed in [12][13][14] for multiobject tracking and traffic flow estimation. The simulations and results are derived using the MatLab Image Processing Toolbox on a work station (GPU) with 128 GB of randomaccess memory and an Intel(R) Xeon(R) Silver 4214 CPU @ 2.19-2.20 GHz.…”

Section: Simulationsmentioning

confidence: 99%

“…Both systems further developed an understanding of the ROI based only on the activities in the present frame and neglected the impact of past events on the ROI. To address this gap, the approach proposed in [14] considered spatiotemporal evaluation for activity mapping. The approach proposed a spatiotemporal relationship between the present and past activities in the frame; however, it did not provide a filter to remove noise or undesirable object detection from the scene.…”

Section: Introductionmentioning

confidence: 99%

“…The approaches in [12][13][14] provide simple models for activity mapping and ROI detection; however, they lack the accuracy of ROI prediction. The methods proposed in [15][16][17][18][19] provide efficient spatiotemporal activity mapping for ROI detection, but at the cost of high computational complexity, and are thus not suitable for systems with low or limited resources.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Spatiotemporal Activity Mapping for Enhanced Multi-Object Detection with Reduced Resource Utilization

Shashank

Sreedevi

2022

Electronics

View full text Add to dashboard Cite

The accuracy of data captured by sensors highly impacts the performance of a computer vision system. To derive highly accurate data, the computer vision system must be capable of identifying critical objects and activities in the field of sensors and reconfiguring the configuration space of the sensors in real time. The majority of modern reconfiguration systems rely on complex computations and thus consume lots of resources. This may not be a problem for systems with a continuous power supply, but it can be a major set-back for computer vision systems employing sensors with limited resources. Further, to develop an appropriate understanding of the scene, the computer vision system must correlate past and present events of the scene captured in the sensor’s field of view (FOV). To address the abovementioned problems, this article provides a simple yet efficient framework for a sensor’s reconfiguration. The framework performs a spatiotemporal evaluation of the scene to generate adaptive activity maps, based on which the sensors are reconfigured. The activity maps contain normalized values assigned to each pixel in the sensor’s FOV, called normalized pixel sensitivity, which represents the impact of activities or events on each pixel in the sensor’s FOV. The temporal relationship between the past and present events is developed by utilizing standard half-width Gaussian distribution. The framework further proposes a federated optical-flow-based filter to determine critical activities in the FOV. Based on the activity maps, the sensors are re-configured to align the center of the sensors to the most sensitive area (i.e., region of importance) of the field. The proposed framework is tested on multiple surveillance and sports datasets and outperforms the contemporary reconfiguration systems in terms of multi-object tracking accuracy (MOTA).

show abstract