Author name cluster

Xiaolong Wu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

10 papers

2 author rows

ICML Conference 2025 Conference Paper

Recent advancements in computer vision have seen Implicit Neural Representations (INR) becoming a dominant representation form for data due to their compactness and expressive power. To solve various vision tasks with INR data, vision networks can either be purely INR-based, but are thereby limited by simplistic operations and performance constraints, or include raster-based methods, which then tend to lose crucial structural information of the INR during the conversion process. To address these issues, we propose DVI, a novel Derivative-based Vision network for INR, capable of handling a variety of vision tasks across various data modalities, while achieving the best performance among the existing methods by incorporating state of the art raster-based methods into a INR based architecture. DVI excels by extracting semantic information from the high order derivative map of the INR, then seamlessly fusing it into a pre-existing raster-based vision network, enhancing its performance with deeper, task-relevant semantic insights. Extensive experiments on five vision tasks across three data modalities demonstrate DVI’s superiority over existing methods. Additionally, our study encompasses comprehensive ablation studies to affirm the efficacy of each element of DVI, the influence of different derivative computation techniques and the impact of derivative orders. Reproducible codes are provided in the supplementary materials.

Details

JBHI Journal 2025 Journal Article

Effects of Different Preprocessing Pipelines on Motor Imagery-Based Brain-Computer Interfaces

Xin Gao
Kai Gui
Xiaolong Wu
Benjamin Metcalfe
Dingguo Zhang

In recent years, brain-computer interfaces (BCIs) leveraging electroencephalography (EEG) signals for the control of external devices have garnered increasing attention. The information transfer rate of BCI has been significantly improved by a lot of cutting-edge methods. The exploration of effective preprocessing in brain-computer interfaces, particularly in terms of identifying suitable preprocessing methods and determining the optimal sequence for their application, remains an area ripe for further investigation. To address this gap, this study explores a range of preprocessing techniques, including but not limited to independent component analysis, surface Laplacian, bandpass filtering, and baseline correction, examining their potential contributions and synergies in the context of BCI applications. In this extensive research, a variety of preprocessing pipelines were rigorously tested across four EEG data sets, all of which were pertinent to motor imagery-based BCIs. These tests incorporated five EEG machine learning models, working in tandem with the preprocessing methods discussed earlier. The study's results highlighted that baseline correction and bandpass filtering consistently provided the most beneficial preprocessing effects. From the perspective of online deployment, after testing and time complexity analysis, this study recommends baseline correction, bandpass filtering and surface Laplace as more suitable for online implementation. An interesting revelation of the study was the enhanced effectiveness of the surface Laplacian algorithm when used alongside algorithms that focus on spatial information. Using appropriate processing algorithms, we can even achieve results (92. 91% and 88. 11%) that exceed the SOTA feature extraction methods in some cases. Such findings are instrumental in offering critical insights for the selection of effective preprocessing pipelines in EEG signal decoding. This, in turn, contributes to the advancement and refinement of brain-computer interface technologies.

Details DOI

AAAI Conference 2024 Conference Paper

Combining Multiple Supervision for Robust Zero-Shot Dense Retrieval

Yan Fang
Qingyao Ai
Jingtao Zhan
Yiqun Liu
Xiaolong Wu
Zhao Cao

Recently, dense retrieval (DR) models, which represent queries and documents with fixed-width vectors and retrieve relevant ones via nearest neighbor search, have drawn increasing attention from the IR community. However, previous studies have shown that the effectiveness of DR critically relies on sufficient training signals, which leads to severe performance degradation when applied in out-of-domain scenarios, where large-scale training data are usually unavailable. To solve this problem, existing studies adopt a data-augmentation-plus-joint-training paradigm to construct weak/pseudo supervisions on the target domain and combine them with the large-scale human annotated data on the source domain to train the DR models. However, they don't explicitly distinguish the data and the supervision signals in the training process and simply assume that the DR models are mighty enough to capture and memorize different domain knowledge and relevance matching patterns without guidance, which, as shown in this paper, is not true. Based on this observation, we propose a Robust Multi-Supervision Combining strategy (RMSC) that decouples the domain and supervision signals by explicitly telling the DR models how the domain data and supervision signals are combined in the training data with specially designed soft tokens. With the extra soft tokens to store the domain-specific and supervision-specific knowledge, RMSC allows the DR models to conduct retrieval based on human-like relevance matching patterns and target-specific language distribution on the target domain without human annotations. Extensive experiments on zero-shot DR benchmarks show that RMSC significantly improves the ranking performance on the target domain compared to strong DR baselines and domain adaptation methods, while being stable during training and can be combined with query generation or second-stage pre-training.

PDF Details DOI

EAAI Journal 2024 Journal Article

Self-organizing broad network with frequency-domain analysis

Honggui Han
Zecheng Tang
Xiaolong Wu
Hongyan Yang
Junfei Qiao

The broad network (BN) based on random feature extraction has fast computational nature. However, it usually suffers from redundant features due to its randomization in dealing with massive samples, which brings the risk of overfitting. To overcome this problem, a self-organizing BN (SO-BN) is proposed in this article. First, several groups of polynomial-based fuzzy rules (P-FRs) are embedded into BN instead of the original feature nodes. Then, P-FR with the form of approximated bell functions has the capability to cope with the uncertainties of feature extraction. Second, a frequency domain parameter calculation algorithm is presented to update the parameters of P-FRs in SO-BN. Different from traditional randomization, P-FRs are shaped with frequency-domain analysis for achieving the representative features of samples. Third, an efficient self-organizing mechanism is built to adjust the structure of the enhancement layer dynamically. Then, the enhancement layer can be expanded and pruned rapidly to reduce the redundant feature as well as improve the performance of SO-BN. Finally, the proposed SO-BN is tested on two benchmark datasets and three real-world engineering applications, especially the prediction of sunspot numbers, electrical output and total phosphorus concentration. The results indicate that SO-BN can achieve superior prediction performance than other models.

Details DOI

AAAI Conference 2024 Conference Paper

Wikiformer: Pre-training with Structured Information of Wikipedia for Ad-Hoc Retrieval

Weihang Su
Qingyao Ai
Xiangsheng Li
Jia Chen
Yiqun Liu
Xiaolong Wu
Shengluan Hou

With the development of deep learning and natural language processing techniques, pre-trained language models have been widely used to solve information retrieval (IR) problems. Benefiting from the pre-training and fine-tuning paradigm, these models achieve state-of-the-art performance. In previous works, plain texts in Wikipedia have been widely used in the pre-training stage. However, the rich structured information in Wikipedia, such as the titles, abstracts, hierarchical heading (multi-level title) structure, relationship between articles, references, hyperlink structures, and the writing organizations, has not been fully explored. In this paper, we devise four pre-training objectives tailored for IR tasks based on the structured knowledge of Wikipedia. Compared to existing pre-training methods, our approach can better capture the semantic knowledge in the training corpus by leveraging the human-edited structured data from Wikipedia. Experimental results on multiple IR benchmark datasets show the superior performance of our model in both zero-shot and fine-tuning settings compared to existing strong retrieval baselines. Besides, experimental results in biomedical and legal domains demonstrate that our approach achieves better performance in vertical domains compared to previous models, especially in scenarios where long text similarity matching is needed. The code is available at https://github.com/oneal2000/Wikiformer.

PDF Details DOI

JBHI Journal 2023 Journal Article

Deep Learning With Convolutional Neural Networks for Motor Brain-Computer Interfaces Based on Stereo-Electroencephalography (SEEG)

Xiaolong Wu
Shize Jiang
Guangye Li
Shengjie Liu
Benjamin Metcalfe
Liang Chen
Dingguo Zhang

Objective: Deep learning based on convolutional neural networks (CNN) has achieved success in brain-computer interfaces (BCIs) using scalp electroencephalography (EEG). However, the interpretation of the so-called ‘black box’ method and its application in stereo-electroencephalography (SEEG)-based BCIs remain largely unknown. Therefore, in this paper, an evaluation is performed on the decoding performance of deep learning methods on SEEG signals. Methods: Thirty epilepsy patients were recruited, and a paradigm including five hand and forearm motion types was designed. Six methods, including filter bank common spatial pattern (FBCSP) and five deep learning methods (EEGNet, shallow and deep CNN, ResNet, and a deep CNN variant named STSCNN), were used to classify the SEEG data. Various experiments were conducted to investigate the effect of windowing, model structure, and the decoding process of ResNet and STSCNN. Results: The average classification accuracy for EEGNet, FBCSP, shallow CNN, deep CNN, STSCNN, and ResNet were 35 $\pm$ 6. 1%, 38 $\pm$ 4. 9%, 60 $\pm$ 3. 9%, 60 $\pm$ 3. 3%, 61 $\pm$ 3. 2%, and 63 $\pm$ 3. 1% respectively. Further analysis of the proposed method demonstrated clear separability between different classes in the spectral domain. Conclusion: ResNet and STSCNN achieved the first- and second-highest decoding accuracy, respectively. The STSCNN demonstrated that an extra spatial convolution layer was beneficial, and the decoding process can be partially interpreted from spatial and spectral perspectives. Significance: This study is the first to investigate the performance of deep learning on SEEG signals. In addition, this paper demonstrated that the so-called ‘black-box’ method can be partially interpreted.

Details DOI

ICRA Conference 2021 Conference Paper

Simultaneous Multi-Level Descriptor Learning and Semantic Segmentation for Domain-Specific Relocalization

Xiaolong Wu
Yiye Chen
Cédric Pradalier
Patricio A. Vela

This paper presents a semi-supervised framework for multi-level description learning aiming for robust and accurate camera relocalization across large perception variations. Our proposed network, namely DLSSNet, simultaneously learns weakly-supervised semantic segmentation and local feature description in the hierarchy. Therefore, the augmented descriptors, trained in an end-to-end manner, provide a more stable high-level representation for local feature dis-ambiguity. To facilitate end-to-end semantic description learning, the descriptor segmentation module is proposed to jointly learn semantic descriptors and cluster centers using standard semantic segmentation loss. We show that our model can be easily fine-tuned for domain-specific usage without any further semantic annotations, instead, requiring only 2D-2D pixel correspondences. The learned descriptors, trained with our proposed pipeline, can boost the cross-season localization performance against other state-of-the-arts.

Details

IROS Conference 2020 Conference Paper

Robust Monocular Edge Visual Odometry through Coarse-to-Fine Data Association

Xiaolong Wu
Patricio A. Vela
Cédric Pradalier

This work describes a monocular visual odometry framework, which exploits the best attributes of edge features for illumination-robust camera tracking, while at the same time ameliorating the performance degradation of edge mapping. In the front-end, an ICP-based edge registration provides robust motion estimation and coarse data association under lighting changes. In the back-end, a novel edge-guided data association pipeline searches for the best photometrically matched points along geometrically possible edges through template matching, so that the matches can be further refined in later bundle adjustment. The core of our proposed data association strategy lies in a point-to-edge geometric uncertainty analysis, which analytically derives (1) a probabilistic search length formula that significantly reduces the search space and (2) a geometric confidence metric for mapping degradation detection based on the predicted depth uncertainty. Moreover, a match confidence based patch size adaption strategy is integrated into our pipeline to reduce matching ambiguity. We present extensive analysis and evaluation of our proposed system on synthetic and real- world benchmark datasets under the influence of illumination changes and large camera motions, where our proposed system outperforms current state-of-art algorithms.

Details

ICRA Conference 2019 Conference Paper

Design and Implementation of Computer Vision based In-Row Weeding System

Xiaolong Wu
Stéphanie Aravecchia
Cédric Pradalier

Autonomous robotic weeding systems in precision farming have demonstrated their full potential to alleviate the current dependency on herbicides or pesticides by introducing selective spraying or mechanical weed removal modules, thus reducing the environmental pollution and improving the sustainability. However, most previous works require fast weed detection system to achieve real-time treatment. In this paper, a novel computer vision based weeding control system is presented, where a non-overlapping multi-camera system is introduced to compensate the indeterminate classification delays, thus allowing for more complicated and advanced detection algorithms, e. g. deep learning based methods. The suitable tracking and control strategies are developed to achieve accurate and robust in-row weed treatment, and the performance of the proposed system is evaluated in different terrain conditions in the presence of various delays.

Details

ICRA Conference 2019 Conference Paper

Illumination Robust Monocular Direct Visual Odometry for Outdoor Environment Mapping

Xiaolong Wu
Cédric Pradalier

Vision-based localization and mapping in outdoor environments is still a challenging issue, which requests significant robustness against various unpredictable illumination changes. In this paper, an illumination-robust direct monocular SLAM system that focuses on modeling outdoor scenery is presented. To deal with global and local lighting changes, such as solar flares, the state-of-art illumination invariant photometric costs for RGB-D and stereo SLAM systems are revisited in the context of their monocular counterpart, where the camera motion and scene structure are jointly optimized with a reasonably poor initialization. Based on our analysis, a combined cost is proposed to achieve a high-precision motion estimation with an improved convergence radius. The proposed system is extensively evaluated on the synthetic and real-world datasets regarding accuracy, robustness, and processing time, where our approach outperforms systems with other costs and state-of-art DSO and ORBSLAM2 systems.

Details