arXiv subjects

Si Zuo

Publications and source records attributed to Si Zuo.

7 recordsLinked to original sources

arXiv2026

RTE-FM-Dehazer: Radiative Transfer Equation Inspired Flow Matching for Real-World Image Dehazing

Single-image dehazing aims to recover a clear scene from a hazy image and is generally formulated as an image-to-image translation task; however, it faces two limitations. Its performance depends heavily on the haze-formation priors embedded in the model. Prevailing methods adopt the Atmospheric Scattering Model (ASM), whose assumptions of single scattering and homogeneous media are often violated, leading to residual haze and color drift. Moreover, large-scale real hazy/clear pairs are impractical to collect, and existing synthesis approaches fail to reproduce the full complexity of natural haze. To address these issues, we present RTE-FM-Dehazer, a novel dehazing approach, together with a scalable data pipeline. Unlike the ASM, the Radiative Transfer Equation (RTE) jointly accounts for both scattering and absorption, naturally accommodating the non-homogeneous, multiple-scattering media that characterize real hazy scenes. Motivated by the structural similarity between the RTE diffusion-absorption term and the ODE in flow matching, we introduce a diffusion-absorption regularizer derived from a reduced RTE, to steer the flow matching trajectory at each step. Next, leveraging modern vision-language models, we build an automated pipeline and release P-HAZE, a dataset of 50000 realistic hazy/clear pairs. Extensive evaluations demonstrate that RTE-FM-Dehazer, trained solely on P-HAZE, effectively eliminates artifacts like residual haze and color drift, exhibits strong cross-domain generalization, and achieves leading results on five real-world dehazing benchmarks.

cs.CV↗

arXiv2026

ImmCOGNITO: Identity Obfuscation in Millimeter-Wave Radar-Based Gesture Recognition for IoT Environments

Millimeter-Wave (mmWave) radar enables camera-free gesture recognition for Internet of Things (IoT) interfaces, with robustness to lighting variations and partial occlusions. However, recent studies reveal that its data can inadvertently encode biometric signatures, raising critical privacy challenges for IoT applications. In particular, we demonstrate that mmWave radar point cloud data can leak identity-related information in the absence of explicit identity labels. To address this risk, we propose {ImmCOGNITO}, a graph-based autoencoder that transforms radar gesture point clouds to preserve gesture-relevant structure while suppressing identity cues. The encoder first constructs a directed graph for each sequence using Temporal Graph KNN. Edges are defined to capture inter-frame temporal dynamics. A message-passing neural network with multi-head self-attention then aggregates local and global spatio-temporal context, and the global max-pooled feature is concatenated with the original features. The decoder then reconstructs a minimally perturbed point cloud that retains gesture discriminative attributes while achieving de-identification. Training jointly optimizes reconstruction, gesture-preservation, and de-identification objectives. Evaluations on two public datasets, PantoRad and MHomeGes, show that ImmCOGNITO substantially reduces identification accuracy while maintaining high gesture recognition performance.

cs.HC↗

arXiv2025

RF-Behavior: A Multimodal Radio-Frequency Dataset for Human Behavior and Emotion Analysis

Recent research has demonstrated the complementary nature of camera-based and inertial data for modeling human gestures, activities, and sentiment. Yet, despite its growing importance for environmental sensing as well as the advance of joint communication and sensing for prospective WiFi and 6G standards, a dataset that integrates these modalities with radio frequency data (radar and RFID) remains rare. We introduce RF-Behavior, a multimodal radio frequency dataset for comprehensive human behavior and emotion analysis. We collected data from 44 participants performing 21 gestures, 10 activities, and 6 sentiment expressions. Data were captured using synchronized sensors, including 13 radars (8 ground-mounted and 5 ceiling-mounted), 6 to 8 RFID tags (attached to each arm) and LoRa. Inertial measurement units (IMUs) and 24 infrared cameras are used to provide precise motion ground truth. RF-Behavior provides a unified multimodal dataset spanning the full spectrum of human behavior -- from brief gestures to activities and emotional states -- enabling research on multi-task learning across motion and emotion recognition. Benchmark results demonstrate that the strategic sensor placement is complementary across modalities, with distinct performance characteristics across different behavioral categories.

cs.DB↗

arXiv2025

LiDARDustX: A LiDAR Dataset for Dusty Unstructured Road Environments

Autonomous driving datasets are essential for validating the progress of intelligent vehicle algorithms, which include localization, perception, and prediction. However, existing datasets are predominantly focused on structured urban environments, which limits the exploration of unstructured and specialized scenarios, particularly those characterized by significant dust levels. This paper introduces the LiDARDustX dataset, which is specifically designed for perception tasks under high-dust conditions, such as those encountered in mining areas. The LiDARDustX dataset consists of 30,000 LiDAR frames captured by six different LiDAR sensors, each accompanied by 3D bounding box annotations and point cloud semantic segmentation. Notably, over 80% of the dataset comprises dust-affected scenes. By utilizing this dataset, we have established a benchmark for evaluating the performance of state-of-the-art 3D detection and segmentation algorithms. Additionally, we have analyzed the impact of dust on perception accuracy and delved into the causes of these effects. The data and further information can be accessed at: https://github.com/vincentweikey/LiDARDustX.

cs.CV↗

arXiv2024

LiSD: An Efficient Multi-Task Learning Framework for LiDAR Segmentation and Detection

With the rapid proliferation of autonomous driving, there has been a heightened focus on the research of lidar-based 3D semantic segmentation and object detection methodologies, aiming to ensure the safety of traffic participants. In recent decades, learning-based approaches have emerged, demonstrating remarkable performance gains in comparison to conventional algorithms. However, the segmentation and detection tasks have traditionally been examined in isolation to achieve the best precision. To this end, we propose an efficient multi-task learning framework named LiSD which can address both segmentation and detection tasks, aiming to optimize the overall performance. Our proposed LiSD is a voxel-based encoder-decoder framework that contains a hierarchical feature collaboration module and a holistic information aggregation module. Different integration methods are adopted to keep sparsity in segmentation while densifying features for query initialization in detection. Besides, cross-task information is utilized in an instance-aware refinement module to obtain more accurate predictions. Experimental results on the nuScenes dataset and Waymo Open Dataset demonstrate the effectiveness of our proposed model. It is worth noting that LiSD achieves the state-of-the-art performance of 83.3% mIoU on the nuScenes segmentation benchmark for lidar-only methods.

cs.CV↗

arXiv2023

Unsupervised Statistical Feature-Guided Diffusion Model for Sensor-based Human Activity Recognition

Human activity recognition (HAR) from on-body sensors is a core functionality in many AI applications: from personal health, through sports and wellness to Industry 4.0. A key problem holding up progress in wearable sensor-based HAR, compared to other ML areas, such as computer vision, is the unavailability of diverse and labeled training data. Particularly, while there are innumerable annotated images available in online repositories, freely available sensor data is sparse and mostly unlabeled. We propose an unsupervised statistical feature-guided diffusion model specifically optimized for wearable sensor-based human activity recognition with devices such as inertial measurement unit (IMU) sensors. The method generates synthetic labeled time-series sensor data without relying on annotated training data. Thereby, it addresses the scarcity and annotation difficulties associated with real-world sensor data. By conditioning the diffusion model on statistical information such as mean, standard deviation, Z-score, and skewness, we generate diverse and representative synthetic sensor data. We conducted experiments on public human activity recognition datasets and compared the method to conventional oversampling and state-of-the-art generative adversarial network methods. Experimental results demonstrate that this can improve the performance of human activity recognition and outperform existing techniques.

eess.SP↗

arXiv2018

A Hierarchical Neural Network for Sequence-to-Sequences Learning

In recent years, the sequence-to-sequence learning neural networks with attention mechanism have achieved great progress. However, there are still challenges, especially for Neural Machine Translation (NMT), such as lower translation quality on long sentences. In this paper, we present a hierarchical deep neural network architecture to improve the quality of long sentences translation. The proposed network embeds sequence-to-sequence neural networks into a two-level category hierarchy by following the coarse-to-fine paradigm. Long sentences are input by splitting them into shorter sequences, which can be well processed by the coarse category network as the long distance dependencies for short sentences is able to be handled by network based on sequence-to-sequence neural network. Then they are concatenated and corrected by the fine category network. The experiments shows that our method can achieve superior results with higher BLEU(Bilingual Evaluation Understudy) scores, lower perplexity and better performance in imitating expression style and words usage than the traditional networks.

cs.CL↗