Publicación:
FAST CNN STEREO DEPTH ESTIMATION THROUGH EMBEDDED GPU DEVICES

Vista sencilla de ítem
dc.creatorCRISTHIAN ALEJANDRO AGUILERA CARRASCO
dc.date2020
dc.date.accessioned2025-01-10T15:16:28Z
dc.date.available2025-01-10T15:16:28Z
dc.date.issued2020
dc.description.abstractCURRENT CNN-BASED STEREO DEPTH ESTIMATION MODELS CAN BARELY RUN UNDER REAL-TIME CONSTRAINTS ON EMBEDDED GRAPHIC PROCESSING UNIT (GPU) DEVICES. MOREOVER, STATE-OF-THE-ART EVALUATIONS USUALLY DO NOT CONSIDER MODEL OPTIMIZATION TECHNIQUES, BEING THAT IT IS UNKNOWN WHAT IS THE CURRENT POTENTIAL ON EMBEDDED GPU DEVICES. IN THIS WORK, WE EVALUATE TWO STATE-OF-THE-ART MODELS ON THREE DIFFERENT EMBEDDED GPU DEVICES, WITH AND WITHOUT OPTIMIZATION METHODS, PRESENTING PERFORMANCE RESULTS THAT ILLUSTRATE THE ACTUAL CAPABILITIES OF EMBEDDED GPU DEVICES FOR STEREO DEPTH ESTIMATION. MORE IMPORTANTLY, BASED ON OUR EVALUATION, WE PROPOSE THE USE OF A U-NET LIKE ARCHITECTURE FOR POSTPROCESSING THE COST-VOLUME, INSTEAD OF A TYPICAL SEQUENCE OF 3D CONVOLUTIONS, DRASTICALLY AUGMENTING THE RUNTIME SPEED OF CURRENT MODELS. IN OUR EXPERIMENTS, WE ACHIEVE REAL-TIME INFERENCE SPEED, IN THE RANGE OF 5-32 MS, FOR 1216 × 368 INPUT STEREO IMAGES ON THE JETSON TX2, JETSON XAVIER, AND JETSON NANO EMBEDDED DEVICES.
dc.formatapplication/pdf
dc.identifier.doi10.3390/s20113249
dc.identifier.issn1424-8220
dc.identifier.issn1424-8220
dc.identifier.urihttps://repositorio.ubiobio.cl/handle/123456789/11236
dc.languagespa
dc.publisherSENSORS
dc.relation.uri10.3390/s20113249
dc.rightsPUBLICADA
dc.titleFAST CNN STEREO DEPTH ESTIMATION THROUGH EMBEDDED GPU DEVICES
dc.title.alternativeESTIMACIÓN RÁPIDA DE LA PROFUNDIDAD ESTÉREO DE CNN A TRAVÉS DE DISPOSITIVOS GPU INTEGRADOS
dc.typeARTÍCULO
dspace.entity.typePublication
ubb.EstadoPUBLICADA
ubb.Otra ReparticionDEPARTAMENTO DE INGENIERIA ELECTRICA Y ELECTRONICA
ubb.SedeCONCEPCIÓN
Archivos
Colecciones
Artículos científicos