[HTML][HTML] Feature-based visual simultaneous localization and mapping: A survey
Visual simultaneous localization and mapping (SLAM) has attracted high attention over the
past few years. In this paper, a comprehensive survey of the state-of-the-art feature-based …
past few years. In this paper, a comprehensive survey of the state-of-the-art feature-based …
fpgaConvNet: Mapping regular and irregular convolutional neural networks on FPGAs
SI Venieris, CS Bouganis - IEEE transactions on neural …, 2018 - ieeexplore.ieee.org
Since neural networks renaissance, convolutional neural networks (ConvNets) have
demonstrated a state-of-the-art performance in several emerging artificial intelligence tasks …
demonstrated a state-of-the-art performance in several emerging artificial intelligence tasks …
[HTML][HTML] Multi-scale fully convolutional network-based semantic segmentation for mobile robot navigation
In computer vision and mobile robotics, autonomous navigation is crucial. It enables the
robot to navigate its environment, which consists primarily of obstacles and moving objects …
robot to navigate its environment, which consists primarily of obstacles and moving objects …
Extracting scientific figures with distantly supervised neural networks
Non-textual components such as charts, diagrams and tables provide key information in
many scientific documents, but the lack of large labeled datasets has impeded the …
many scientific documents, but the lack of large labeled datasets has impeded the …
Pop-up slam: Semantic monocular plane slam for low-texture environments
Existing simultaneous localization and mapping (SLAM) algorithms are not robust in
challenging low-texture environments because there are only few salient features. The …
challenging low-texture environments because there are only few salient features. The …
[HTML][HTML] Obstacle avoidance strategy for mobile robot based on monocular camera
This research paper proposes a real-time obstacle avoidance strategy for mobile robots with
a monocular camera. The approach uses a binary semantic segmentation FCN-VGG-16 to …
a monocular camera. The approach uses a binary semantic segmentation FCN-VGG-16 to …
Semantic 3D occupancy mapping through efficient high order CRFs
Semantic 3D mapping can be used for many applications such as robot navigation and
virtual interaction. In recent years, there has been great progress in semantic segmentation …
virtual interaction. In recent years, there has been great progress in semantic segmentation …
A survey on deep learning methods for robot vision
J Ruiz-del-Solar, P Loncomilla, N Soto - arXiv preprint arXiv:1803.10862, 2018 - arxiv.org
Deep learning has allowed a paradigm shift in pattern recognition, from using hand-crafted
features together with statistical classifiers to using general-purpose learning procedures for …
features together with statistical classifiers to using general-purpose learning procedures for …
Kaolin: A pytorch library for accelerating 3d deep learning research
KM Jatavallabhula, E Smith, JF Lafleche… - arXiv preprint arXiv …, 2019 - arxiv.org
We present Kaolin, a PyTorch library aiming to accelerate 3D deep learning research.
Kaolin provides efficient implementations of differentiable 3D modules for use in deep …
Kaolin provides efficient implementations of differentiable 3D modules for use in deep …
并行注意力机制在图像语义分割中的应用.
张汉， 张德祥， 陈鹏， 章军… - Journal of Computer …, 2022 - search.ebscohost.com
在卷积神经网络中融入注意力机制越来越成为语义分割强化特征学习的重要方法.
提出了一种融合了局部注意力和全局注意力的卷积神经网络. 输入图像经主干网络的特征提取 …
提出了一种融合了局部注意力和全局注意力的卷积神经网络. 输入图像经主干网络的特征提取 …