Bio-inspired heterogeneous architecture for real-time pedestrian detection applications

Maggiani, Luca; Bourrasset, Cédric; Quinton, Jean‐Charles; Berry, François; Sérot, Jocelyn

doi:10.1007/s11554-016-0581-3

Cited by 15 publications

(8 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides the people counting scenario, we also evaluate the effects of intelligence partitioning in the node energy consumption for the smart camera nodes introduced by Imran et al [8] and Maggiani et al [9]. The energy consumption results regarding the computation are, respectively, based on the Xilinx Power Estimation tool and PowerPlay Early Power Estimator tool [48] with a 20% error margin.…”

Section: Pedestrian and Particle Detectionmentioning

confidence: 99%

“…The LTE Cat. 1 technology is an exception to the group [8]; c pedestrian detection [9] performance, as it meets the delay requirements for data rates up to 11, 264 bytes per frame.…”

Section: Delay and Channel Usagementioning

confidence: 99%

“…We investigate how the allocation of image-processing tasks on embedded hardware, software or cloud affects the overall node energy consumption, for a people counting scenario. Furthermore, we implement this approach to the scenarios presented by [8] and [9] to compare the energy efficiency of this method with their results. The aim of this work is not only to provide an energy-efficient design for a people counting scenario, but mainly to explore the trade-off between computation and communication energy for data-intensive IoT scenarios.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Communication and computation inter-effects in people counting using intelligence partitioning

Shallari

Krug

O’Nils

2020

J Real-Time Image Proc

View full text Add to dashboard Cite

The rapid development of the Internet of Things is affecting the requirements towards wireless vision sensor networks (WVSN). Future smart camera architectures require battery-operated devices to facilitate deployment for scenarios such as industrial monitoring, environmental monitoring and smart city, consequently imposing constraints on the node energy consumption. This paper provides an analysis of the inter-effects between computation and communication energy for a smart camera node. Based on a people counting scenario, we evaluate the trade-off for the node energy consumption with different processing configurations of the image processing tasks, and several communication technologies. The results indicate that the optimal partition between the smart camera node and remote processing is with background modelling, segmentation, morphology and binary compression implemented in the smart camera, supported by Bluetooth Low Energy (BLE) version 5 technologies. The comparative assessment of these results with other implementation scenarios underlines the energy efficiency of this approach. This work changes pre-conceptions regarding design space exploration in WVSN, motivating further investigation regarding the inclusion of intermediate processing layers between the node and the cloud to interlace low-power configurations of communication and processing architectures.

show abstract

Section: Pedestrian and Particle Detectionmentioning

confidence: 99%

“…The LTE Cat. 1 technology is an exception to the group [8]; c pedestrian detection [9] performance, as it meets the delay requirements for data rates up to 11, 264 bytes per frame.…”

Section: Delay and Channel Usagementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Communication and computation inter-effects in people counting using intelligence partitioning

Shallari

Krug

O’Nils

2020

J Real-Time Image Proc

View full text Add to dashboard Cite

show abstract

“…Also, DNF models do not need to be directly plugged on sensory inputs. Therefore, they are not in principle limited to the detection of stimuli defined by a single lateral connectivity profile (for an instance of complex preprocessing to abstract from raw input images, see Maggiani, Bourrasset, Quinton, Berry, & Sérot, 2016). Also, complex transformations and projections can be considered, for instance using alog-polar representation of the visual field (Taouali, Goffart, Alexandre, & Rougier, 2015) or relying on self-organizing maps (Lefort, Boniface, & Girau, 2011).…”

Section: Dynamic Neural Fieldsmentioning

confidence: 99%

A unified dynamic neural field model of goal directed eye movements

Quinton

Goffart

2018

Connection Science

View full text Add to dashboard Cite

Primates heavily rely on their visual system, which exploits signals of graded precision based on the eccentricity of the target in the visual field. The interactions with the environment involve actively selecting and focusing on visual targets or regions of interest, instead of contemplating an omnidirectional visual flow. Eye-movements specifically allow foveating targets and track their motion. Once a target is brought within the central visual field, eye-movements are usually classified into catch-up saccades (jumping from one orientation or fixation to another) and smooth pursuit (continuously tracking a target with low velocity). Building on existing dynamic neural field equations, we introduce a novel model that incorporates internal projections to better estimate the current target location (associated to a peak of activity). Such estimate is then used to trigger an eye movement, leading to qualitatively different behaviors depending on the dynamics of the whole oculomotor system: 1) fixational eye-movements due to small variations in the weights of projections when the target is stationary, 2) interceptive and catch-up saccades when peaks build and relax on the neural field, 3) smooth pursuit when the peak stabilizes near the center of the field, the system reaching a fixed point attractor. Learning is nevertheless required for tracking a rapidly moving target, and the proposed model thus replicates recent results in the monkey, in which repeated exercise permits the maintenance of the target within in the central visual field at its current (here-and-now) location, despite the delays involved in transmitting retinal signals to the oculomotor neurons.

show abstract

“…The second article entitled ''Bio-inspired heterogeneous architecture for real-time pedestrian detection applications'' by Maggiani et al [9] presents a custom-tailored hardware/software architecture for real-time pedestrian detection as can be used as ADAS. More specifically, the authors tightly couple a computationally intensive histogram of oriented gradient (HOG) pipeline implemented in an FPGA with a bio-inspired spatiotemporal filter running on a CPU.…”

Section: Heterogeneous Real-time Image Processingmentioning

confidence: 99%