Orcc

Yviquel, Hervé; Lorence, Antoine; Jerbi, Khaled; Cocherel, Gildas; Sanchez, Alexandre; Raulet, Mickaël

doi:10.1145/2502081.2502231

Cited by 77 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Bezati et al [3] and Pelcat et al [34], top-down approaches use the data-flow model of the application to ease the parallelism exploration to target heterogeneous multi-core architectures. Other authors use a high-level programming language that implements a data-flow model of computation that is to be automatically transformed into a valid hardware implementation or multi-core software, see [41,49]. Some works rely on the UML model with its architecture specific extensions (MARTE, UML 2.0 see [25] … ) Fig.…”

Section: Hw/sw Co-design Methodologiesmentioning

confidence: 99%

See 1 more Smart Citation

HW/SW co-design of a visual SLAM application

Piat

Fillatreau

Tortei

et al. 2018

J Real-Time Image Proc

View full text Add to dashboard Cite

Vision-based advanced driver assistance systems (ADAS), appeared in the 2000s, are increasingly integrated on-board mass-produced vehicles, as off-the-shelf low-cost cameras are now available. But ADAS implement most of the timespecific and basic functionalities such as lane departure or control of the distance to other vehicles. Integrating accurate localization and mapping functionalities meeting the constraints of ADAS (high-throughput, low-consumption, and smalldesign footprint) would pave the way towards obstacle detection, identification and tracking on-board vehicles at potential high speed. While the SLAM problem has been widely addressed by the robotics community, very few embedded operational implementations can be found, and they do not meet the ADAS-related constraints. In this paper, we implement the first 3D monocular EKF-SLAM chain on a heterogeneous architecture, on a single System on Chip (SoC), meeting these constraints. In order to do so, we picked up a standard co-design method (Shaout et al. Specification and modeling of hw/sw co-design for heterogeneous embedded systems, 2009) and adapted it to the implementation of potentially any of such complex processing chains. The refined method encompasses a hardware-in-the-loop approach allowing to progressively integrate hardware accelerators on the basis of a systematic rule. We also have designed original hardware accelerators for all the image processing functions involved, and for some algebraic operations involved in the filtering process.

show abstract

Section: Hw/sw Co-design Methodologiesmentioning

confidence: 99%

“…The flow does not specify how the designer improves the knowledge along the design iterations. • fully automated design flows (see for example [41,49]) somehow fail to capture the full extent of the designer's knowledge.…”

Section: Proposed Co-design Methodologymentioning

confidence: 99%

HW/SW co-design of a visual SLAM application

Piat

Fillatreau

Tortei

et al. 2018

J Real-Time Image Proc

View full text Add to dashboard Cite

show abstract

“…General purpose dataflow models like ORCC [20], Apache NiFi [21] and Apache Spark [22] allow programmers to connect modules in a flexible manner to help compose diverse applications. In contrast, we give a high-level dataflow composition to meet the specific needs of tracking applications, and focus on specific implementations of these modules based on advances in DL/CV models, and uniquely, control the distributed tracking logic through a custom module.…”

Section: Domain-specific Programming Modelmentioning

confidence: 99%

“…General purpose dataflow models such as ORCC [20] and Apache NiFi [21] give programmers the flexibility to compose complex applications using logic blocks, often providing pre-defined blocks and a graphical UI. These are then compiled and executed within a runtime environment.…”

Section: Big Data Platforms and Dslmentioning

confidence: 99%

A Scalable Platform for Distributed Object Tracking Across a Many-Camera Network

Khochare

Krishnan

Simmhan

2021

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

Advances in deep neural networks (DNN) and computer vision (CV) algorithms have made it feasible to extract meaningful insights from large-scale deployments of urban cameras. Tracking an object of interest across the camera network in near real-time is a canonical problem. However, current tracking platforms have two key limitations: 1) They are monolithic, proprietary and lack the ability to rapidly incorporate sophisticated tracking models, and 2) They are less responsive to dynamism across wide-area computing resources that include edge, fog, and cloud abstractions. We address these gaps using Anveshak, a runtime platform for composing and coordinating distributed tracking applications. It provides a domain-specific dataflow programming model to intuitively compose a tracking application, supporting contemporary CV advances like query fusion and re-identification, and enabling dynamic scoping of the camera network's search space to avoid wasted computation. We also offer tunable batching and data-dropping strategies for dataflow blocks deployed on distributed resources to respond to network and compute variability. These balance the tracking accuracy, its real-time performance, and the active camera-set size. We illustrate the concise expressiveness of the programming model for four tracking applications. Our detailed experiments for a network of 1000 camera-feeds on modest resources exhibit the tunable scalability, performance, and quality trade-offs enabled by our dynamic tracking, batching, and dropping strategies.

show abstract

“…About three decades ago, a number of toolkits were developed to support multimedia and video content, including X Toolkit [38], Ttoolkit [25], MET++ [2], Video Widgets [22] and VideoScheme [36]. Since then, work has shifted towards online platforms: Hop [50] offers web-based multimedia programming similar to Orcc [61] that builds on a dataflow programming model. NUBOMEDIA [18] provides an entire suite based on WebRTC for programmable cloud-based editing.…”

Section: Research On Video Interactionmentioning

confidence: 99%