Adrian Munteanu
YOU?
Author Swipe
View article: Non-Uniform Entropy-Constrained L∞ Quantization for Sparse and Irregular Sources
Non-Uniform Entropy-Constrained L∞ Quantization for Sparse and Irregular Sources Open
Near-lossless coding schemes traditionally rely on uniform quantization to control the maximum absolute error (L∞ norm) of residual signals, often assuming a parametric model for the source distribution. This paper introduces a novel desig…
View article: Air Conditioning Systems in Vehicles: Approaches and Challenges
Air Conditioning Systems in Vehicles: Approaches and Challenges Open
Automotive air conditioning systems improve passenger comfort and safety while keeping pace with changing environmental and technological requirements. This review evaluates the historical development, technological progress, and future tr…
View article: ReferGPT: Towards Zero-Shot Referring Multi-Object Tracking
ReferGPT: Towards Zero-Shot Referring Multi-Object Tracking Open
Tracking multiple objects based on textual queries is a challenging task that requires linking language understanding with object association across frames. Previous works typically train the whole process end-to-end or integrate an additi…
View article: Lossless and Near-Lossless L-Infinite Compression of Depth Video Data
Lossless and Near-Lossless L-Infinite Compression of Depth Video Data Open
The acquisition of depth information sensorial data is critically important in medical applications, such as the monitoring of the elderly or the extraction of human biometrics. In such applications, compressing the stream of depth video d…
View article: Non-Uniform Voxelisation for Point Cloud Compression
Non-Uniform Voxelisation for Point Cloud Compression Open
Point cloud compression is essential for the efficient storage and transmission of 3D data in various applications, such as virtual reality, autonomous driving, and 3D modelling. Most existing compression methods employ voxelisation, all o…
View article: Lossless and Near-Lossless L-Infinite Compression of Depth Video Data
Lossless and Near-Lossless L-Infinite Compression of Depth Video Data Open
Depth information acquisition is critically important in medical applications, such as monitoring of elderly or human biometrics extraction. In such applications, compressing the stream of depth video data plays an important role due to ba…
View article: Non-Uniform Voxelisation for Point Cloud Compression
Non-Uniform Voxelisation for Point Cloud Compression Open
Point cloud compression is essential for the efficient storage and transmission of 3D data in various applications, such as virtual reality, autonomous driving, and 3D modelling. Most existing compression methods employ voxelisation, all o…
View article: HybridTrack: A Hybrid Approach for Robust Multi-Object Tracking
HybridTrack: A Hybrid Approach for Robust Multi-Object Tracking Open
The evolution of Advanced Driver Assistance Systems (ADAS) has increased the need for robust and generalizable algorithms for multi-object tracking. Traditional statistical model-based tracking methods rely on predefined motion models and …
View article: Where are You? Unscented Particle Filter for Single Range Relative Pose Estimation in Unobservable Motion Using UWB and VIO
Where are You? Unscented Particle Filter for Single Range Relative Pose Estimation in Unobservable Motion Using UWB and VIO Open
Real-time relative pose (RP) estimation is a cornerstone for effective multi-agent collaboration. When conventional global positioning infrastructure such as GPS is unavailable, the use of Ultra-Wideband (UWB) technology on each agent p…
View article: Strada-LLM: Graph LLM for traffic prediction
Strada-LLM: Graph LLM for traffic prediction Open
Traffic forecasting is pivotal for intelligent transportation systems, where accurate and interpretable predictions can significantly enhance operational efficiency and safety. A key challenge stems from the heterogeneity of traffic condit…
View article: ProtoSeg: A Prototype-Based Point Cloud Instance Segmentation Method
ProtoSeg: A Prototype-Based Point Cloud Instance Segmentation Method Open
3D instance segmentation is crucial for obtaining an understanding of a point cloud scene. This paper presents a novel neural network architecture for performing instance segmentation on 3D point clouds. We propose to jointly learn coeffic…
View article: RESSCAL3D++: Joint Acquisition and Semantic Segmentation of 3D Point Clouds
RESSCAL3D++: Joint Acquisition and Semantic Segmentation of 3D Point Clouds Open
3D scene understanding is crucial for facilitating seamless interaction\nbetween digital devices and the physical world. Real-time capturing and\nprocessing of the 3D scene are essential for achieving this seamless\nintegration. While exis…
View article: GINTRIP: Interpretable Temporal Graph Regression using Information bottleneck and Prototype-based method
GINTRIP: Interpretable Temporal Graph Regression using Information bottleneck and Prototype-based method Open
Deep neural networks (DNNs) have demonstrated remarkable performance across various domains, but their inherent complexity makes them challenging to interpret. This is especially true for temporal graph regression tasks due to the complex …
View article: Efficient and Scalable Point Cloud Generation with Sparse Point-Voxel Diffusion Models
Efficient and Scalable Point Cloud Generation with Sparse Point-Voxel Diffusion Models Open
We propose a novel point cloud U-Net diffusion architecture for 3D generative modeling capable of generating high-quality and diverse 3D shapes while maintaining fast generation times. Our network employs a dual-branch architecture, combin…
View article: LAM3D: Leveraging Attention for Monocular 3D Object Detection
LAM3D: Leveraging Attention for Monocular 3D Object Detection Open
Since the introduction of the self-attention mechanism and the adoption of the Transformer architecture for Computer Vision tasks, the Vision Transformer-based architectures gained a lot of popularity in the field, being used for tasks suc…
View article: RT-GS2: Real-Time Generalizable Semantic Segmentation for 3D Gaussian Representations of Radiance Fields
RT-GS2: Real-Time Generalizable Semantic Segmentation for 3D Gaussian Representations of Radiance Fields Open
Gaussian Splatting has revolutionized the world of novel view synthesis by achieving high rendering performance in real-time. Recently, studies have focused on enriching these 3D representations with semantic information for downstream tas…
View article: Embodied greenhouse gas emissions of buildings—Machine learning approach for early stage prediction
Embodied greenhouse gas emissions of buildings—Machine learning approach for early stage prediction Open
Observations made in architecture and engineering practices have highlighted the need to access buildings embodied greenhouse gas (GHG) emissions at early stages as well as a general understanding of the impacts of design decisions. This r…
View article: DeepKalPose: An enhanced deep‐learning Kalman filter for temporally consistent monocular vehicle pose estimation
DeepKalPose: An enhanced deep‐learning Kalman filter for temporally consistent monocular vehicle pose estimation Open
This paper presents DeepKalPose, a novel approach for enhancing temporal consistency in monocular vehicle pose estimation applied on video through a deep‐learning‐based Kalman filter. By integrating a bi‐directional Kalman filter strategy …
View article: A UWB-Ego-Motion Particle Filter for Indoor Pose Estimation of a Ground Robot Using a Moving Horizon Hypothesis
A UWB-Ego-Motion Particle Filter for Indoor Pose Estimation of a Ground Robot Using a Moving Horizon Hypothesis Open
Ultra-wideband (UWB) has gained increasing interest for providing real-time positioning to robots in GPS-denied environments. For a robot to act on this information, it also requires its heading. This is, however, not provided by UWB. To o…
View article: DeepKalPose: An Enhanced Deep-Learning Kalman Filter for Temporally Consistent Monocular Vehicle Pose Estimation
DeepKalPose: An Enhanced Deep-Learning Kalman Filter for Temporally Consistent Monocular Vehicle Pose Estimation Open
In this paper, we introduce an innovative temporal consistency enhancement approach, which enables image-based models on video data by leveraging a deep-learning-based Kalman Filter. More specifically, we propose a novel Bi-direction Kalma…
View article: Joint prototype and coefficient prediction for 3D instance segmentation
Joint prototype and coefficient prediction for 3D instance segmentation Open
3D instance segmentation is crucial for applications demanding comprehensive 3D scene understanding. Here, a novel method is introduced that simultaneously learns coefficients and prototypes. Employing an overcomplete sampling strategy, th…
View article: PCGen: A Fully Parallelizable Point Cloud Generative Model
PCGen: A Fully Parallelizable Point Cloud Generative Model Open
Generative models have the potential to revolutionize 3D extended reality. A primary obstacle is that augmented and virtual reality need real-time computing. Current state-of-the-art point cloud random generation methods are not fast enoug…
View article: ExpPoint-MAE: Better Interpretability and Performance for Self-Supervised Point Cloud Transformers
ExpPoint-MAE: Better Interpretability and Performance for Self-Supervised Point Cloud Transformers Open
In this paper we delve into the properties of transformers, attained through self-supervision, in the point cloud domain. Specifically, we evaluate the effectiveness of Masked Autoencoding as a pretraining scheme, and explore Momentum Cont…
View article: Selecting and Combining UWB Localization Algorithms: Insights and Recommendations From a Multi-Metric Benchmark
Selecting and Combining UWB Localization Algorithms: Insights and Recommendations From a Multi-Metric Benchmark Open
Ultra-wideband (UWB) localization has emerged in GPS-denied environments as a crucial facilitator for diverse industries, including logistics, healthcare applications, and societal domains. Despite notable progress, UWB algorithms from the…
View article: Table of Contents
Table of Contents Open
A novel sensorless control of a hybrid excitation
View article: KD-Net: Continuous-Keystroke-Dynamics-Based Human Identification from RGB-D Image Sequences
KD-Net: Continuous-Keystroke-Dynamics-Based Human Identification from RGB-D Image Sequences Open
Keystroke dynamics is a soft biometric based on the assumption that humans always type in uniquely characteristic manners. Previous works mainly focused on analyzing the key press or release events. Unlike these methods, we explored a nove…
View article: GPU Rasterization-Based 3D LiDAR Simulation for Deep Learning
GPU Rasterization-Based 3D LiDAR Simulation for Deep Learning Open
High-quality data are of utmost importance for any deep-learning application. However, acquiring such data and their annotation is challenging. This paper presents a GPU-accelerated simulator that enables the generation of high-quality, pe…
View article: RESSCAL3D: Resolution Scalable 3D Semantic Segmentation of Point Clouds
RESSCAL3D: Resolution Scalable 3D Semantic Segmentation of Point Clouds Open
While deep learning-based methods have demonstrated outstanding results in\nnumerous domains, some important functionalities are missing. Resolution\nscalability is one of them. In this work, we introduce a novel architecture,\ndubbed RESS…