20.500.12556/RUL-184133 Do it once concatenating the image pair for a single pass feature extraction in stereo depth sensing In the field of stereo depth sensing, modern research predominantly prioritizes accuracy, yet inference speed remains a critical bottleneck for practical, real-time applications on resource-constrained platforms. Existing acceleration approaches often rely on lighter network architectures or runtime-specific optimizations, which may require architectural redesign, platform-specific tuning, or accuracy trade-offs. However, a common inefficiency remains in many stereo pipelines: feature extraction is typically performed using two separate forward passes, one for the left image and one for the right, even though both passes use the same network weights. We address this redundancy by concatenating the left and right images into a single combined tensor, enabling feature extraction in one batched pass while preserving the original network architecture. By reducing feature extraction time by up to 48.4%, our results demonstrate that this method accelerates the overall inference rate by 10% to 39% on average on Nvidia V100 and up to 28.4% on edge device, depending on the model architecture. This speedup is achieved at the expense of only a moderate increase in runtime memory consumption, while retaining the original accuracy. Because the method does not alter the core stereo network, it can be applied as a plug-and-play enhancement to both existing and newly developed stereo matching models. stereo vision depth estimation real-time inference feature extraction edge computing stereo vid ocena globine sklepanje v realnem času ekstrakcija značilnosti računalništvo na robu true false true Angleški jezik Slovenski jezik Članek v reviji 2026-06-30 10:15:32 2026-06-30 10:15:38 2026-07-02 03:28:59 0000-00-00 00:00:00 2026 0 0 19 str. iss. 12, art. 3919 Vol. 26 2026 0000-00-00 Zaloznikova NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.932:004.8 1424-8220 283047939 Do_it_once_Sensors.pdf Do_it_once_Sensors.pdf 1 2CEC69A6A609EEE2662022C2A04DF056 a3fc2fa1effac47900890462dbd46043976b2d73e2987bc251b24ff6c8209a7a 8c653a56-745c-11f1-9b0d-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=236646 https://www.mdpi.com/1424-8220/26/12/3919 1 c77e22fa-745b-11f1-9b0d-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=236644 Fakulteta za elektrotehniko 0 0 0