RT info:eu-repo/semantics/article
T1 Robustifying semantic perception of traversability across wearable RGB-depth cameras
A1 Bergasa Pascual, Luis Miguel
A1 Romera Carmena, Eduardo
A1 Wang, Kaiwei
A1 Yang, Kailun
K1 Electrónica
K1 Electronics
AB Semantic segmentation represents a promising means to unify different detection tasks, especially pixelwise traversability perception as the fundamental enabler in robotic vision systems aiding upper-levelnavigational applications. However, major research efforts are being put into earning marginal accuracy increments on semantic segmentation benchmarks, without assuring the robustness of real-timesegmenters to be deployed in assistive cognition systems for the visually impaired. In this paper, weexplore in a comparative study across four perception systems, including a pair of commercial smartglasses, a customized wearable prototype and two portable RGB-Depth (RGB-D) cameras that are beingintegrated in the next generation of navigation assistance devices. More concretely, we analyze the gapbetween the concepts of “accuracy” and “robustness” on the critical traversability-related semantic sceneunderstanding. A cluster of efficient deep architectures is proposed, which are built using spatial factorizations, hierarchical dilations and pyramidal representations. Based on these architectures, this researchdemonstrates the augmented robustness of semantically traversable area parsing against the variationsof environmental conditions in diverse RGB-D observations, and sensorial factors such as illumination,imaging quality, field of view and detectable depth range.
PB OSA
SN 1559-128X
YR 2019
FD 2019-04
LK http://hdl.handle.net/10017/43167
UL http://hdl.handle.net/10017/43167
LA eng
NO Ministerio de Economía y Competitividad
DS MINDS@UW
RD 26-abr-2024