Related papers: Gleanmer: A 6 mW SoC for Real-Time 3D Gaussian Occ…

Graph convolutional network (GCN), an emerging algorithm for graph computing, has achieved promising performance in graphstructure tasks. To achieve acceleration for data-intensive and sparse graph computing, ASICs such as GCNAX have been…

3D Gaussian splatting (3DGS) has recently emerged as an alternative representation that leverages a 3D Gaussian-based representation and introduces an approximated volumetric rendering, achieving very fast rendering speed and promising…

Spectrum sharing, as an approach to significantly improve spectrum efficiency in the era of 6th generation mobile networks (6G), has attracted extensive attention. Radio Environment Map (REM) based low-complexity spectrum sharing is widely…

Most of the existing robotic exploration schemes use occupancy grid representations and geometric targets known as frontiers. The occupancy grid representation relies on the assumption of independence between grid cells and ignores…

Understanding the geometric and semantic structure of environments is essential for embodied navigation and reasoning. Existing semantic mapping methods trade off between explicit geometry and multi-scale semantics, and lack a native…

We propose an efficient and scalable method for incrementally building a dense, semantically annotated 3D map in real-time. The proposed method assigns class probabilities to each region, not each element (e.g., surfel and voxel), of the 3D…

Voxel grids are a minimal and efficient environment representation that is used for robot motion planning in numerous tasks. Many state-of-the-art planning algorithms use voxel grids composed of free, occupied and unknown voxels. In this…

Long Short-Term Memory (LSTM) neural networks have penetrated healthcare applications where real-time requirements and edge computing capabilities are essential. Gait analysis that detects abnormal steps to prevent patients from falling is…

This paper proposes a real-time multi-plane segmentation method based on GPU-accelerated high-resolution 3D voxel mapping for legged robot locomotion. Existing online planar mapping approaches struggle to balance accuracy and computational…

3D semantic occupancy prediction is crucial for autonomous driving. While multi-modal fusion improves accuracy over vision-only methods, it typically relies on computationally expensive dense voxel or BEV tensors. We present Gau-Occ, a…

Graph convolutional networks (GCNs) have been introduced to effectively process non-euclidean graph data. However, GCNs incur large amounts of irregularity in computation and memory access, which prevents efficient use of traditional neural…

Optimizing robot poses and the map simultaneously has been shown to provide more accurate SLAM results. However, for non-feature based SLAM approaches, directly optimizing all the robot poses and the whole map will greatly increase the…

Simultaneous localization and mapping (SLAM) plays a vital role in mapping unknown spaces and aiding autonomous navigation. Virtually all state-of-the-art solutions today for 2D SLAM are designed for dense and accurate sensors such as laser…

While there have been many studies on hardware acceleration for deep learning on images, there has been a rather limited focus on accelerating deep learning applications involving graphs. The unique characteristics of graphs, such as the…

On-device machine learning (ODML) enables powerful edge applications, but power consumption remains a key challenge for resource-constrained devices. To address this, developers often face a trade-off between model accuracy and power…

Robotic perception requires the modeling of both 3D geometry and semantics. Existing methods typically focus on estimating 3D bounding boxes, neglecting finer geometric details and struggling to handle general, out-of-vocabulary objects. 3D…

Full-graph training of graph neural networks (GNNs) is widely used as it enables direct validation of algorithmic improvements by preserving complete neighborhood information. However, it typically requires multiple GPUs or servers,…

3D Gaussian Splatting (3DGS) achieves an appealing balance between rendering quality and efficiency, but relies on approximating 3D Gaussians as 2D projections--an assumption that degrades accuracy, especially under generic large…

The advent of 3D Gaussian Splatting has revolutionized graphics rendering by delivering high visual quality and fast rendering speeds. However, training large-scale scenes at high quality remains challenging due to the substantial memory…

In this paper, we develop a high-dimensional map building technique that incorporates raw pixelated semantic measurements into the map representation. The proposed technique uses Gaussian Processes (GPs) multi-class classification for map…