Xinyi Wu

Tingqin Lai, Xiaolin Liang, Yi Zhu et al.

Recently, there has been increased attention towards 3D imaging using single-pixel single-photon detection (also known as temporal data) due to its potential advantages in terms of cost and power efficiency. However, to eliminate the symmetry blur in the reconstructed images, a fixed background is required. This paper proposes a fusion-data-based 3D imaging method that utilizes a single-pixel single-photon detector and a millimeter-wave radar to capture temporal histograms of a scene from multiple perspectives. Subsequently, the 3D information can be reconstructed from the one-dimensional fusion temporal data by using Artificial Neural Network (ANN). Both the simulation and experimental results demonstrate that our fusion method effectively eliminates symmetry blur and improves the quality of the reconstructed images.

Xinyi Wu, Yanhao Jia, Luwei Xiao et al.

In AI-facilitated teaching, leveraging various query styles to interpret abstract educational content is crucial for delivering effective and accessible learning experiences. However, existing retrieval systems predominantly focus on natural text-image matching and lack the capacity to address the diversity and ambiguity inherent in real-world educational scenarios. To address this limitation, we develop a lightweight and efficient multi-modal retrieval module, named Uni-Retrieval, which extracts query-style prototypes and dynamically matches them with tokens from a continually updated Prompt Bank. This Prompt Bank encodes and stores domain-specific knowledge by leveraging a Mixture-of-Expert Low-Rank Adaptation (MoE-LoRA) module and can be adapted to enhance Uni-Retrieval's capability to accommodate unseen query types at test time. To enable natural language educational content generation, we integrate the original Uni-Retrieval with a compact instruction-tuned language model, forming a complete retrieval-augmented generation pipeline named Uni-RAG. Given a style-conditioned query, Uni-RAG first retrieves relevant educational materials and then generates human-readable explanations, feedback, or instructional content aligned with the learning objective. Experimental results on SER and other multi-modal benchmarks show that Uni-RAG outperforms baseline retrieval and RAG systems in both retrieval accuracy and generation quality, while maintaining low computational cost. Our framework provides a scalable, pedagogically grounded solution for intelligent educational systems, bridging retrieval and generation to support personalized, explainable, and efficient learning assistance across diverse STEM scenarios.

Ruixuan Sun, Xinyi Wu, Avinash Akella et al.

In the past decade, deep learning (DL) models have gained prominence for their exceptional accuracy on benchmark datasets in recommender systems (RecSys). However, their evaluation has primarily relied on offline metrics, overlooking direct user perception and experience. To address this gap, we conduct a human-centric evaluation case study of four leading DL-RecSys models in the movie domain. We test how different DL-RecSys models perform in personalized recommendation generation by conducting survey study with 445 real active users. We find some DL-RecSys models to be superior in recommending novel and unexpected items and weaker in diversity, trustworthiness, transparency, accuracy, and overall user satisfaction compared to classic collaborative filtering (CF) methods. To further explain the reasons behind the underperformance, we apply a comprehensive path analysis. We discover that the lack of diversity and too much serendipity from DL models can negatively impact the consequent perceived transparency and personalization of recommendations. Such a path ultimately leads to lower summative user satisfaction. Qualitatively, we confirm with real user quotes that accuracy plus at least one other attribute is necessary to ensure a good user experience, while their demands for transparency and trust can not be neglected. Based on our findings, we discuss future human-centric DL-RecSys design and optimization strategies.

Seyed Mohammad Hassan Erfani, Zhenyao Wu, Xinyi Wu et al.

Vision-based semantic segmentation of waterbodies and nearby related objects provides important information for managing water resources and handling flooding emergency. However, the lack of large-scale labeled training and testing datasets for water-related categories prevents researchers from studying water-related issues in the computer vision field. To tackle this problem, we present ATLANTIS, a new benchmark for semantic segmentation of waterbodies and related objects. ATLANTIS consists of 5,195 images of waterbodies, as well as high quality pixel-level manual annotations of 56 classes of objects, including 17 classes of man-made objects, 18 classes of natural objects and 21 general classes. We analyze ATLANTIS in detail and evaluate several state-of-the-art semantic segmentation networks on our benchmark. In addition, a novel deep neural network, AQUANet, is developed for waterbody semantic segmentation by processing the aquatic and non-aquatic regions in two different paths. AQUANet also incorporates low-level feature modulation and cross-path modulation for enhancing feature representation. Experimental results show that the proposed AQUANet outperforms other state-of-the-art semantic segmentation networks on ATLANTIS. We claim that ATLANTIS is the largest waterbody image dataset for semantic segmentation providing a wide range of water and water-related classes and it will benefit researchers of both computer vision and water resources engineering.

Xinyi Wu, Zhenyao Wu, Hao Guo et al.

Semantic segmentation of nighttime images plays an equally important role as that of daytime images in autonomous driving, but the former is much more challenging due to poor illuminations and arduous human annotations. In this paper, we propose a novel domain adaptation network (DANNet) for nighttime semantic segmentation without using labeled nighttime image data. It employs an adversarial training with a labeled daytime dataset and an unlabeled dataset that contains coarsely aligned day-night image pairs. Specifically, for the unlabeled day-night image pairs, we use the pixel-level predictions of static object categories on a daytime image as a pseudo supervision to segment its counterpart nighttime image. We further design a re-weighting strategy to handle the inaccuracy caused by misalignment between day-night image pairs and wrong predictions of daytime images, as well as boost the prediction accuracy of small objects. The proposed DANNet is the first one stage adaptation framework for nighttime semantic segmentation, which does not train additional day-night image transfer models as a separate pre-processing stage. Extensive experiments on Dark Zurich and Nighttime Driving datasets show that our method achieves state-of-the-art performance for nighttime semantic segmentation.

Zhihao Liu, Hui Yin, Xinyi Wu et al.

Shadow removal is a computer-vision task that aims to restore the image content in shadow regions. While almost all recent shadow-removal methods require shadow-free images for training, in ECCV 2020 Le and Samaras introduces an innovative approach without this requirement by cropping patches with and without shadows from shadow images as training samples. However, it is still laborious and time-consuming to construct a large amount of such unpaired patches. In this paper, we propose a new G2R-ShadowNet which leverages shadow generation for weakly-supervised shadow removal by only using a set of shadow images and their corresponding shadow masks for training. The proposed G2R-ShadowNet consists of three sub-networks for shadow generation, shadow removal and refinement, respectively and they are jointly trained in an end-to-end fashion. In particular, the shadow generation sub-net stylises non-shadow regions to be shadow ones, leading to paired data for training the shadow-removal sub-net. Extensive experiments on the ISTD dataset and the Video Shadow Removal dataset show that the proposed G2R-ShadowNet achieves competitive performances against the current state of the arts and outperforms Le and Samaras' patch-based shadow-removal method.

Xinyi Wu, Kartik Balkumar, Qi Luo et al.

Real-time parking occupancy information is critical for a parking management system to facilitate drivers to park more efficiently. Recent advances in connected and automated vehicle technologies enable sensor-equipped cars (probe cars) to detect and broadcast available parking spaces when driving through parking lots. In this paper, we evaluate the impact of market penetration of probe cars on the system performance, and investigate different parking guidance policies to improve the data acquisition process. We adopt a simulation-based approach to impose four policies on an off- street parking lot influencing the behavior of probe cars to park in assigned parking spaces. This in turn effects the scanning route and the parking space occupancy estimations. The last policy we propose is a near-optimal guidance strategy that maximizes the information gain of posteriors. The results suggest that an efficient information gathering policy can compensate for low penetration of connected and automated vehicles. We also highlight the policy trade-off that occur while attempting to maximize information gain through explorations and improve assignment accuracy through exploitations. Our results can assist urban policy makers in designing and managing smart parking systems.

Qi Luo, Romesh Saigal, Robert Hampshire et al.

Real-time parking occupancy information is valuable for guiding drivers' searching for parking spaces. Recently many parking detection systems using range-based on-vehicle sensors are invented, but they disregard the practical difficulty of obtaining access to raw sensory data which are required for any feature-based algorithm. In this paper, we focus on a system using short-range radars (SRR) embedded in Advanced Driver Assistance System (ADAS) to collect occupancy information, and broadcast it through a connected vehicle network. The challenge that the data transmitted through ADAS unit has been encoded to sparse points is overcome by a statistical method instead of feature extractions. We propose a two-step classification algorithm combining Mean-Shift clustering and Support Vector Machine to analyze SRR-GPS data, and evaluate it through field experiments. The results show that the average Type I error rate for off-street parking is $15.23 \%$ and for on-street parking is $32.62\%$. In both cased the Type II error rates are less than $20 \%$. Bayesian updating can recursively improve the mapping results. This paper can provide a comprehensive method to elevate automotive sensors for the parking detection function.