2026-04-13T07:26:11Zhttps://recercat.cat/oai/request
oai:recercat.cat:2117/913512026-02-07T09:08:07Zcom_2072_1033col_2072_452950
RECERCAT author Oro Garcia, David author Fernandez Tena, Carles author Martorell Bofill, Xavier author Hernando Pericás, Francisco Javier 2016 With the emergence of GPU computing, deep neural networks have become a widely used technique for advancing research in the field of image and speech processing. In the context of object and event detection, slidingwindow classifiers require to choose the best among all positively discriminated candidate windows. In this paper, we introduce the first GPU-based non-maximum suppression (NMS) algorithm for embedded GPU architectures. The obtained results show that the proposed parallel algorithm reduces the NMS latency by a wide margin when compared to CPUs, even clocking the GPU at 50% of its maximum frequency on an NVIDIA Tegra K1. In this paper, we show results for object detection in images. The proposed technique is directly applicable to speech segmentation tasks such as speaker diarization.Peer ReviewedPostprint (published version) Restricted access - publisher's policy Àrees temàtiques de la UPC::Informàtica::Sistemes d'informació Àrees temàtiques de la UPC::Informàtica Embedded computer systems Information display systems Embedded systems Graphics processing units Parallel algorithms Work-efficient parallel nonmaximum suppression Embedded GPU architectures Image processing Speech processing Deep neural networks NMS latency Positively discriminated candidate windows Parallel algorithm NVIDIA Tegra JC1 Speech segmentation tasks Speaker diarization Sistemes incrustats (Informàtica) Visualització (Informàtica) Work-efficient parallel non-maximum suppression for embedded GPU architectures Conference report