Source author record

Guoqiang Li

Guoqiang Li appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Catalog footprint

What is connected

14works
14topics
4close collaborators

Actions

Connect this record

Log in to claim

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

14 published item(s)

preprint2026arXiv

OT-Drive: Out-of-Distribution Off-Road Traversable Area Segmentation via Optimal Transport

Reliable traversable area segmentation in unstructured environments is critical for planning and decision-making in autonomous driving. However, existing data-driven approaches often suffer from degraded segmentation performance in out-of-distribution (OOD) scenarios, consequently impairing downstream driving tasks. To address this issue, we propose OT-Drive, an Optimal Transport--driven multi-modal fusion framework. The proposed method formulates RGB and surface normal fusion as a distribution transport problem. Specifically, we design a novel Scene Anchor Generator (SAG) to decompose scene information into the joint distribution of weather, time-of-day, and road type, thereby constructing semantic anchors that can generalize to unseen scenarios. Subsequently, we design an innovative Optimal Transport-based multi-modal fusion module (OT Fusion) to transport RGB and surface normal features onto the manifold defined by the semantic anchors, enabling robust traversable area segmentation under OOD scenarios. Experimental results demonstrate that our method achieves 95.16% mIoU on ORFD OOD scenarios, outperforming prior methods by 6.35%, and 89.79% mIoU on cross-dataset transfer tasks, surpassing baselines by 13.99%.These results indicate that the proposed model can attain strong OOD generalization with only limited training data, substantially enhancing its practicality and efficiency for real-world deployment.

preprint2026arXiv

Teacher-Aware Evolution of Heuristic Programs from Learned Optimization Policies

LLM-based automatic heuristic design has shown promise for generating executable heuristics for combinatorial optimization, but existing methods mainly rely on delayed endpoint performance. We propose a \emph{teacher-aware evolutionary framework} that uses independently trained learned optimization policies as behavioral teachers. Instead of deploying or imitating the teacher, our method queries it on states visited by candidate heuristic programs and uses its action preferences as local feedback for evolution. The resulting search discovers static executable heuristics guided by both task performance and teacher-derived behavioral signals. Experiments on scheduling, routing, and graph optimization benchmarks show that our method improves over performance-driven LLM heuristic evolution baselines while requiring no neural inference at deployment. These results suggest that learned optimization policies can be repurposed as behavioral feedback sources for automatic heuristic discovery.

preprint2020arXiv

A Survey on Unknown Presentation Attack Detection for Fingerprint

Fingerprint recognition systems are widely deployed in various real-life applications as they have achieved high accuracy. The widely used applications include border control, automated teller machine (ATM), and attendance monitoring systems. However, these critical systems are prone to spoofing attacks (a.k.a presentation attacks (PA)). PA for fingerprint can be performed by presenting gummy fingers made from different materials such as silicone, gelatine, play-doh, ecoflex, 2D printed paper, 3D printed material, or latex. Biometrics Researchers have developed Presentation Attack Detection (PAD) methods as a countermeasure to PA. PAD is usually done by training a machine learning classifier for known attacks for a given dataset, and they achieve high accuracy in this task. However, generalizing to unknown attacks is an essential problem from applicability to real-world systems, mainly because attacks cannot be exhaustively listed in advance. In this survey paper, we present a comprehensive survey on existing PAD algorithms for fingerprint recognition systems, specifically from the standpoint of detecting unknown PAD. We categorize PAD algorithms, point out their advantages/disadvantages, and future directions for this area.

preprint2020arXiv

Morphing Attack Detection -- Database, Evaluation Platform and Benchmarking

Morphing attacks have posed a severe threat to Face Recognition System (FRS). Despite the number of advancements reported in recent works, we note serious open issues such as independent benchmarking, generalizability challenges and considerations to age, gender, ethnicity that are inadequately addressed. Morphing Attack Detection (MAD) algorithms often are prone to generalization challenges as they are database dependent. The existing databases, mostly of semi-public nature, lack in diversity in terms of ethnicity, various morphing process and post-processing pipelines. Further, they do not reflect a realistic operational scenario for Automated Border Control (ABC) and do not provide a basis to test MAD on unseen data, in order to benchmark the robustness of algorithms. In this work, we present a new sequestered dataset for facilitating the advancements of MAD where the algorithms can be tested on unseen data in an effort to better generalize. The newly constructed dataset consists of facial images from 150 subjects from various ethnicities, age-groups and both genders. In order to challenge the existing MAD algorithms, the morphed images are with careful subject pre-selection created from the contributing images, and further post-processed to remove morphing artifacts. The images are also printed and scanned to remove all digital cues and to simulate a realistic challenge for MAD algorithms. Further, we present a new online evaluation platform to test algorithms on sequestered data. With the platform we can benchmark the morph detection performance and study the generalization ability. This work also presents a detailed analysis on various subsets of sequestered data and outlines open challenges for future directions in MAD research.

preprint2020arXiv

Object Detection for Graphical User Interface: Old Fashioned or Deep Learning or a Combination?

Detecting Graphical User Interface (GUI) elements in GUI images is a domain-specific object detection task. It supports many software engineering tasks, such as GUI animation and testing, GUI search and code generation. Existing studies for GUI element detection directly borrow the mature methods from computer vision (CV) domain, including old fashioned ones that rely on traditional image processing features (e.g., canny edge, contours), and deep learning models that learn to detect from large-scale GUI data. Unfortunately, these CV methods are not originally designed with the awareness of the unique characteristics of GUIs and GUI elements and the high localization accuracy of the GUI element detection task. We conduct the first large-scale empirical study of seven representative GUI element detection methods on over 50k GUI images to understand the capabilities, limitations and effective designs of these methods. This study not only sheds the light on the technical challenges to be addressed but also informs the design of new GUI element detection methods. We accordingly design a new GUI-specific old-fashioned method for non-text GUI element detection which adopts a novel top-down coarse-to-fine strategy, and incorporate it with the mature deep learning model for GUI text detection.Our evaluation on 25,000 GUI images shows that our method significantly advances the start-of-the-art performance in GUI element detection.

preprint2020arXiv

Unblind Your Apps: Predicting Natural-Language Labels for Mobile GUI Components by Deep Learning

According to the World Health Organization(WHO), it is estimated that approximately 1.3 billion people live with some forms of vision impairment globally, of whom 36 million are blind. Due to their disability, engaging these minority into the society is a challenging problem. The recent rise of smart mobile phones provides a new solution by enabling blind users' convenient access to the information and service for understanding the world. Users with vision impairment can adopt the screen reader embedded in the mobile operating systems to read the content of each screen within the app, and use gestures to interact with the phone. However, the prerequisite of using screen readers is that developers have to add natural-language labels to the image-based components when they are developing the app. Unfortunately, more than 77% apps have issues of missing labels, according to our analysis of 10,408 Android apps. Most of these issues are caused by developers' lack of awareness and knowledge in considering the minority. And even if developers want to add the labels to UI components, they may not come up with concise and clear description as most of them are of no visual issues. To overcome these challenges, we develop a deep-learning based model, called LabelDroid, to automatically predict the labels of image-based buttons by learning from large-scale commercial apps in Google Play. The experimental results show that our model can make accurate predictions and the generated labels are of higher quality than that from real Android developers.

preprint2015arXiv

Energy flow structuring in the focused field

We propose an iterative method of energy flow shaping in the focal region with the amplitude, phase and polarization modulation of incident light. By using an iterative optimization based on the diffraction calculation with help of the fast Fourier transform, we can tailor the polarization and phase structure in the focal plane. By appropriate design of the polarization and phase gradients, arbitrary energy flow including spin and orbital parts can be designed and tailored independently. The capability of energy flow structuring is demonstrated by the measurement of the Stokes parameters and self-interference pattern. This provides a novel method to control the vectorial feature of the focal volume.

preprint2015arXiv

Independent and simultaneous tailoring of amplitude, phase, and complete polarization of vector beams

We present an approach that enables complete control over the amplitude, phase and arbitrary polarization state on the Poincaré sphere of an optical beam in a 4-f system with a spatial light modulator (SLM). The beams can be constructed from a coaxial superposition of x- and y-linearly polarized light, each carrying structured amplitude profile and phase distributions by using an amplitude-modulated mask imposed on the SLM. The amplitude, phase and polarization distribution of vector beams with four free parameters can be tailored independently and simultaneously by the SLM.

preprint2015arXiv

Iso-oriented monolayer α-MoO3(010) films epitaxially grown on SrTiO3(001)

The ability to synthesis well-ordered two-dimensional materials under ultra-high vacuum and directly characterize them by other techniques in-situ can greatly advance our current understanding on their physical and chemical properties. In this paper, we demonstrate that iso-oriented α-MoO3 films with as low as single monolayer thickness can be reproducibly grown on SrTiO3(001) (STO) substrates by molecular beam epitaxy ( (010)MoO3 || (001)STO, [100]MoO3 || [100]STO or [010]STO) through a self-limiting process. While one in-plane lattice parameter of the MoO3 is very close to that of the SrTiO3 (aMoO3 = 3.96 Å, aSTO = 3.905 Å), the lattice mismatch along other direction is large (~5%, cMoO3 = 3.70 Å), which leads to relaxation as clearly observed from the splitting of streaks in reflection high-energy electron diffraction (RHEED) patterns. A narrow range in the growth temperature is found to be optimal for the growth of monolayer α-MoO3 films. Increasing deposition time will not lead to further increase in thickness, which is explained by a balance between deposition and thermal desorption due to the weak van der Waals force between α-MoO3 layers. Lowering growth temperature after the initial iso-oriented α-MoO3 monolayer leads to thicker α-MoO3(010) films with excellent crystallinity.

preprint2015arXiv

Phase transition from focal conic to cubic smectic blue phase in partially fluorinated cyano-phenyl alkyl benzoate ester doped with ultrahigh twisting power chiral dopant

Blue phase liquid crystal (BPLC) has important applications in adaptive lenses and phase modulators due to its polarization-independent property. During our efforts for development of the new materials, we found a novel phenomenology of phase transition, from focal conic smectic to smectic blue phase in a partially fluorinated cyanophenyl alkyl benzoate ester based nematic liquid crystal (LCM-5773) doped by ultra-high twisting power [H.T.P~160 um^-1] chiral dopant (R5011/3 wt%). Polarized optical microscopy (POM) investigations revealed focal conic and fan-shaped textures typical for columnar mesophases. These focal conic domains (FCDs) are squeezed under electric field and finally at a critical electric field they undergo a dark state. When the electric field is withdrawn, the FCDs are regrown in a one dimensional array with smaller domain size. Interestingly, we have observed the domain size of the FCDs can grow several times by decreasing the cooling rate (0.02 degrees(C)/min.) ten times without any change in the phase sequence. In blue phase (BP), we have observed curved platelet texture and grain boundaries filled by small platelets, which is completely different from conventional cholesteric BP. The blue phase platelet size (PLS) also increases significantly at low cooling rates. The thermal control of FCD and PLS size has increasing demand for the construction of devices with optimal performances.

preprint2015arXiv

Space-variant polarized Airy beam

We experimentally generate an Airy beam with polarization structure while keeping its original amplitude and phase profile intact. This class of Airy beam preserves the acceleration properties. By monitoring their initial polarization structure we have provided insight concerning the self-healing mechanism of Airy beams. We investigate both theoretically and experimentally the self-healing polarization properties of the space-variant polarized Airy beams. Amplitude as well as the polarization structure tends to reform during propagation in spite of the severe truncation of the beam by finite apertures.

preprint2015arXiv

Theoretical analysis of nanoparticle-induced homeotropic alignment in nematic liquid crystals

A theoretical analysis of homeotropic alignment induced by nanoparticles (NPs) in a nematic liquid crystal (NLC) sample cell is presented. It is found that such alignment on the surface of a NP causes a change in the orientation of the molecular director near the surface, which in turn induces variations in the elastic constants and free energy. The induced NLC properties allow coupling between nearby NPs, mediated by the NLC molecules. The rotation of the coupled NPs close to the substrate tends to induce a long-range orientation of the NLC molecular director, leading to modification in the alignment at the interface of NLC and substrate which induces the orientation from homogeneous (planar) to homeotropic (vertical) in the bulk material.

preprint2015arXiv

ZenLDA: An Efficient and Scalable Topic Model Training System on Distributed Data-Parallel Platform

This paper presents our recent efforts, zenLDA, an efficient and scalable Collapsed Gibbs Sampling system for Latent Dirichlet Allocation training, which is thought to be challenging that both data parallelism and model parallelism are required because of the Big sampling data with up to billions of documents and Big model size with up to trillions of parameters. zenLDA combines both algorithm level improvements and system level optimizations. It first presents a novel CGS algorithm that balances the time complexity, model accuracy and parallelization flexibility. The input corpus in zenLDA is represented as a directed graph and model parameters are annotated as the corresponding vertex attributes. The distributed training is parallelized by partitioning the graph that in each iteration it first applies CGS step for all partitions in parallel, followed by synchronizing the computed model each other. In this way, both data parallelism and model parallelism are achieved by converting them to graph parallelism. We revisited the tradeoff between system efficiency and model accuracy and presented approximations such as unsynchronized model, sparse model initialization and "converged" token exclusion. zenLDA is built on GraphX in Spark that provides distributed data abstraction (RDD) and expressive APIs to simplify the programming efforts and simultaneously hides the system complexities. This enables us to implement other CGS algorithm with a few lines of code change. To better fit in distributed data-parallel framework and achieve comparable performance with contemporary systems, we also presented several system level optimizations to push the performance limit. zenLDA was evaluated it against web-scale corpus, and the result indicates that zenLDA can achieve about much better performance than other CGS algorithm we implemented, and simultaneously achieve better model accuracy.