arXiv subjects

Feng Lu

Publications and source records attributed to Feng Lu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Gaze Archive: Enhancing Human Memory through Active Visual Logging on Smart Glasses

People today are overwhelmed by massive amounts of information, leading to cognitive overload and memory burden. Traditional visual memory augmentation methods are either effortful and disruptive or fail to align with user intent. To address these limitations, we propose Gaze Archive, a novel visual memory enhancement paradigm through active logging on smart glasses. It leverages human gaze as a natural attention indicator, enabling both intent-precise capture and effortless-and-unobtrusive interaction. To implement Gaze Archive, we develop GAHMA, a technical framework that enables compact yet intent-aligned memory encoding and intuitive memory recall based on natural language queries. Quantitative experiments on our newly constructed GAVER dataset show that GAHMA achieves more intent-precise logging than non-gaze baselines. Through extensive user studies in both laboratory and real-world scenarios, we compare Gaze Archive with other existing memory augmentation methods. Results demonstrate its advantages in perceived effortlessness, unobtrusiveness and overall preference, showing strong potential for real-world deployment.

cs.HC↗

arXiv2025

Learning Procedural-aware Video Representations through State-Grounded Hierarchy Unfolding

Learning procedural-aware video representations is a key step towards building agents that can reason about and execute complex tasks. Existing methods typically address this problem by aligning visual content with textual descriptions at the task and step levels to inject procedural semantics into video representations. However, due to their high level of abstraction, 'task' and 'step' descriptions fail to form a robust alignment with the concrete, observable details in visual data. To address this, we introduce 'states', i.e., textual snapshots of object configurations, as a visually-grounded semantic layer that anchors abstract procedures to what a model can actually see. We formalize this insight in a novel Task-Step-State (TSS) framework, where tasks are achieved via steps that drive transitions between observable states. To enforce this structure, we propose a progressive pre-training strategy that unfolds the TSS hierarchy, forcing the model to ground representations in states while associating them with steps and high-level tasks. Extensive experiments on the COIN and CrossTask datasets show that our method outperforms baseline models on multiple downstream tasks, including task recognition, step recognition, and next step prediction. Ablation studies show that introducing state supervision is a key driver of performance gains across all tasks. Additionally, our progressive pretraining strategy proves more effective than standard joint training, as it better enforces the intended hierarchical structure.

cs.CV↗

arXiv2025

CST Anti-UAV: A Thermal Infrared Benchmark for Tiny UAV Tracking in Complex Scenes

The widespread application of Unmanned Aerial Vehicles (UAVs) has raised serious public safety and privacy concerns, making UAV perception crucial for anti-UAV tasks. However, existing UAV tracking datasets predominantly feature conspicuous objects and lack diversity in scene complexity and attribute representation, limiting their applicability to real-world scenarios. To overcome these limitations, we present the CST Anti-UAV, a new thermal infrared dataset specifically designed for Single Object Tracking (SOT) in Complex Scenes with Tiny UAVs (CST). It contains 220 video sequences with over 240k high-quality bounding box annotations, highlighting two key properties: a significant number of tiny-sized UAV targets and the diverse and complex scenes. To the best of our knowledge, CST Anti-UAV is the first dataset to incorporate complete manual frame-level attribute annotations, enabling precise evaluations under varied challenges. To conduct an in-depth performance analysis for CST Anti-UAV, we evaluate 20 existing SOT methods on the proposed dataset. Experimental results demonstrate that tracking tiny UAVs in complex environments remains a challenge, as the state-of-the-art method achieves only 35.92% state accuracy, much lower than the 67.69% observed on the Anti-UAV410 dataset. These findings underscore the limitations of existing benchmarks and the need for further advancements in UAV tracking research. The CST Anti-UAV benchmark is about to be publicly released, which not only fosters the development of more robust SOT methods but also drives innovation in anti-UAV systems.

cs.CV↗

arXiv2025

Evolution of correlated electronic states of La2NiO4 under hydrostatic pressure

We elucidate the electronic structure and quantum many-body instabilities of the monolayer nickelate La2NiO4 under hydrostatic pressure using a combination of density functional theory, dynamical mean-field theory (DFT+DMFT), and random phase approximation (RPA). Our DFT+DMFT calculations reveal non-Fermi-liquid behavior and coherence loss near the Fermi level at low pressures, driven by strong electron correlations within the Ni-e_g orbital manifold, which is analogous to the low-energy electronic properties observed in La3Ni2O7. However, multi-orbital spin susceptibility analysis demonstrates an exceptionally suppressed critical Stoner parameter U_c (about 0.4~0.7 eV), indicating robust magnetic order that dominates the ground state and precludes superconductivity in the pristine system. Below U_c, superconducting instabilities exhibit a pressure-driven symmetry transition: the d_(x2-y2)-wave pairing prevails at ambient and low pressure, while the s+g-wave symmetry occurs above 75 GPa. This transition is attributed to pressure-induced self-doping effect. The high-angular-momentum g-wave component incurs significant energetic penalties, rendering high-Tc superconductivity unlikely. We conclude that the absence of superconductivity in La2NiO4 arises from its robust intrinsic magnetism and the unfavorable pairing symmetry under pressure, suggesting that alternative routes-such as chemical doping or epitaxial strain-are necessary to suppress magnetism and unlock superconducting states.

cond-mat.str-el↗

arXiv2025

High-Winding-Number Zero-Energy Edge States in Rhombohedral-Stacked Su-Schrieffer-Heeger Multilayers

We study the topological properties of rhombohedral-stacked N-layer Su-Schrieffer-Heeger networks with interlayer coupling. We find that these systems exhibit $2N$-fold degenerate zero-energy edge states with winding number $W=N$, providing a direct route to high-winding-number topological phases where $W$ equals the layer number. Using effective Hamiltonian theory and Zak phase calculations, we demonstrate that the winding number scales linearly with $N$ through a layer-by-layer topological amplification mechanism. We introduce the Wigner entropy as a novel detection method for these edge states, showing that topological boundary states exhibit significantly enhanced Wigner entropy compared to bulk states. Our results establish rhombohedral stacking as a systematic approach for engineering high-winding-number topological insulators with potential applications in quantum information processing.

cond-mat.dis-nn↗

arXiv2025

SelaVPR++: Towards Seamless Adaptation of Foundation Models for Efficient Place Recognition

Recent studies show that the visual place recognition (VPR) method using pre-trained visual foundation models can achieve promising performance. In our previous work, we propose a novel method to realize seamless adaptation of foundation models to VPR (SelaVPR). This method can produce both global and local features that focus on discriminative landmarks to recognize places for two-stage VPR by a parameter-efficient adaptation approach. Although SelaVPR has achieved competitive results, we argue that the previous adaptation is inefficient in training time and GPU memory usage, and the re-ranking paradigm is also costly in retrieval latency and storage usage. In pursuit of higher efficiency and better performance, we propose an extension of the SelaVPR, called SelaVPR++. Concretely, we first design a parameter-, time-, and memory-efficient adaptation method that uses lightweight multi-scale convolution (MultiConv) adapters to refine intermediate features from the frozen foundation backbone. This adaptation method does not back-propagate gradients through the backbone during training, and the MultiConv adapter facilitates feature interactions along the spatial axes and introduces proper local priors, thus achieving higher efficiency and better performance. Moreover, we propose an innovative re-ranking paradigm for more efficient VPR. Instead of relying on local features for re-ranking, which incurs huge overhead in latency and storage, we employ compact binary features for initial retrieval and robust floating-point (global) features for re-ranking. To obtain such binary features, we propose a similarity-constrained deep hashing method, which can be easily integrated into the VPR pipeline. Finally, we improve our training strategy and unify the training protocol of several common training datasets to merge them for better training of VPR models. Extensive experiments show that ......

cs.CV↗

arXiv2025

In the Eye of MLLM: Benchmarking Egocentric Video Intent Understanding with Gaze-Guided Prompting

The emergence of advanced multimodal large language models (MLLMs) has significantly enhanced AI assistants' ability to process complex information across modalities. Recently, egocentric videos, by directly capturing user focus, actions, and context in an unified coordinate, offer an exciting opportunity to enable proactive and personalized AI user experiences with MLLMs. However, existing benchmarks overlook the crucial role of gaze as an indicator of user intent. To address this gap, we introduce EgoGazeVQA, an egocentric gaze-guided video question answering benchmark that leverages gaze information to improve the understanding of longer daily-life videos. EgoGazeVQA consists of gaze-based QA pairs generated by MLLMs and refined by human annotators. Our experiments reveal that existing MLLMs struggle to accurately interpret user intentions. In contrast, our gaze-guided intent prompting methods significantly enhance performance by integrating spatial, temporal, and intent-related cues. We further conduct experiments on gaze-related fine-tuning and analyze how gaze estimation accuracy impacts prompting effectiveness. These results underscore the value of gaze for more personalized and effective AI assistants in egocentric settings. Project page: https://taiyi98.github.io/projects/EgoGazeVQA

cs.CV↗

arXiv2025

LSVOS 2025 Challenge Report: Recent Advances in Complex Video Object Segmentation

This report presents an overview of the 7th Large-scale Video Object Segmentation (LSVOS) Challenge held in conjunction with ICCV 2025. Besides the two traditional tracks of LSVOS that jointly target robustness in realistic video scenarios: Classic VOS (VOS), and Referring VOS (RVOS), the 2025 edition features a newly introduced track, Complex VOS (MOSEv2). Building upon prior insights, MOSEv2 substantially increases difficulty, introducing more challenging but realistic scenarios including denser small objects, frequent disappear/reappear events, severe occlusions, adverse weather and lighting, etc., pushing long-term consistency and generalization beyond curated benchmarks. The challenge retains standard ${J}$, $F$, and ${J\&F}$ metrics for VOS and RVOS, while MOSEv2 adopts ${J\&\dot{F}}$ as the primary ranking metric to better evaluate objects across scales and disappearance cases. We summarize datasets and protocols, highlight top-performing solutions, and distill emerging trends, such as the growing role of LLM/MLLM components and memory-aware propagation, aiming to chart future directions for resilient, language-aware video segmentation in the wild.

cs.CV↗

arXiv2025

Enhancing Sa2VA for Referent Video Object Segmentation: 2nd Solution for 7th LSVOS RVOS Track

Referential Video Object Segmentation (RVOS) aims to segment all objects in a video that match a given natural language description, bridging the gap between vision and language understanding. Recent work, such as Sa2VA, combines Large Language Models (LLMs) with SAM~2, leveraging the strong video reasoning capability of LLMs to guide video segmentation. In this work, we present a training-free framework that substantially improves Sa2VA's performance on the RVOS task. Our method introduces two key components: (1) a Video-Language Checker that explicitly verifies whether the subject and action described in the query actually appear in the video, thereby reducing false positives; and (2) a Key-Frame Sampler that adaptively selects informative frames to better capture both early object appearances and long-range temporal context. Without any additional training, our approach achieves a J&F score of 64.14% on the MeViS test set, ranking 2nd place in the RVOS track of the 7th LSVOS Challenge at ICCV 2025.

cs.CV↗

arXiv2025

Pseudo-Label Enhanced Cascaded Framework: 2nd Technical Report for LSVOS 2025 VOS Track

Complex Video Object Segmentation (VOS) presents significant challenges in accurately segmenting objects across frames, especially in the presence of small and similar targets, frequent occlusions, rapid motion, and complex interactions. In this report, we present our solution for the LSVOS 2025 VOS Track based on the SAM2 framework. We adopt a pseudo-labeling strategy during training: a trained SAM2 checkpoint is deployed within the SAM2Long framework to generate pseudo labels for the MOSE test set, which are then combined with existing data for further training. For inference, the SAM2Long framework is employed to obtain our primary segmentation results, while an open-source SeC model runs in parallel to produce complementary predictions. A cascaded decision mechanism dynamically integrates outputs from both models, exploiting the temporal stability of SAM2Long and the concept-level robustness of SeC. Benefiting from pseudo-label training and cascaded multi-model inference, our approach achieves a J\&F score of 0.8616 on the MOSE test set -- +1.4 points over our SAM2Long baseline -- securing the 2nd place in the LSVOS 2025 VOS Track, and demonstrating strong robustness and accuracy in long, complex video segmentation scenarios.

cs.CV↗

arXiv2025

Wigner distribution, Wigner entropy, and Anomalous Transport of a Generalized Aubry-André model

We investigate generalized Aubry-André models featuring tunable quasidisordered potentials and a mobility edge that separates extended and localized states, with critical states for the mobility edge confirmed through finite-size scaling analysis. Numerical results demonstrate that extended, critical, and localized states can be distinguished via their phase-space representations, particularly the Wigner distribution. The associated Wigner entropy, derived from this distribution, peaks at the critical state, enabling precise localization of the mobility edge. Additionally, wave-packet dynamics reveal anomalous transport behaviors, including superdiffusion and subdiffusion, bridging ballistic transport and the absence of diffusion.

cond-mat.dis-nn↗

arXiv2025

Lighting the Night with Generative Artificial Intelligence

The visible light reflectance data from geostationary satellites is crucial for meteorological observations and plays an important role in weather monitoring and forecasting. However, due to the lack of visible light at night, it is impossible to conduct continuous all-day weather observations using visible light reflectance data. This study pioneers the use of generative diffusion models to address this limitation. Based on the multi-band thermal infrared brightness temperature data from the Advanced Geostationary Radiation Imager (AGRI) onboard the Fengyun-4B (FY4B) geostationary satellite, we developed a high-precision visible light reflectance generative model, called Reflectance Diffusion (RefDiff), which enables 0.47~μ\mathrm{m}, 0.65~μ\mathrm{m}, and 0.825~μ\mathrm{m} bands visible light reflectance generation at night. Compared to the classical models, RefDiff not only significantly improves accuracy through ensemble averaging but also provides uncertainty estimation. Specifically, the SSIM index of RefDiff can reach 0.90, with particularly significant improvements in areas with complex cloud structures and thick clouds. The model's nighttime generation capability was validated using VIIRS nighttime product, demonstrating comparable performance to its daytime counterpart. In summary, this research has made substantial progress in the ability to generate visible light reflectance at night, with the potential to expand the application of nighttime visible light data.

cs.CV↗

arXiv2025

When Homogeneous Systems Meet Dissipation and Disorder

We investigate the localization and topological properties of the non-equilibrium steady state (NESS) in a one-dimensional homogeneous system. Our results demonstrate that, despite the absence of disorder in the initial system, the NESS can exhibit localization under bond dissipation. These dissipation-driven localization and delocalization phenomena are clearly distinguished using Wigner distributions. Furthermore, we find that the initial localization characteristics of the system significantly influence the localization properties of the NESS. Drawing upon the concept of Bose-Einstein condensate broadening in cold-atom experiments, along with experimental data, we systematically characterize the impact of bond dissipation and disorder on the localization and topological properties of the NESS. The phase diagram reveals that the NESS can be topologically non-trivial even when the initial system is topologically trivial, and that the topological triviality of the initial system strengthens the topological non-triviality of the NESS. This work provides new insights into the localization and topological phase transitions in homogeneous systems induced by bond dissipation and disorder.

cond-mat.dis-nn↗

arXiv2025

Towards spatial computing: recent advances in multimodal natural interaction for XR headsets

With the widespread adoption of Extended Reality (XR) headsets, spatial computing technologies are gaining increasing attention. Spatial computing enables interaction with virtual elements through natural input methods such as eye tracking, hand gestures, and voice commands, thus placing natural human-computer interaction at its core. While previous surveys have reviewed conventional XR interaction techniques, recent advancements in natural interaction, particularly driven by artificial intelligence (AI) and large language models (LLMs), have introduced new paradigms and technologies. In this paper, we review research on multimodal natural interaction for wearable XR, focusing on papers published between 2022 and 2024 in six top venues: ACM CHI, UIST, IMWUT (Ubicomp), IEEE VR, ISMAR, and TVCG. We classify and analyze these studies based on application scenarios, operation types, and interaction modalities. This analysis provides a structured framework for understanding how researchers are designing advanced natural interaction techniques in XR. Based on these findings, we discuss the challenges in natural interaction techniques and suggest potential directions for future research. This review provides valuable insights for researchers aiming to design natural and efficient interaction systems for XR, ultimately contributing to the advancement of spatial computing.

cs.HC↗

arXiv2025

EDTformer: An Efficient Decoder Transformer for Visual Place Recognition

Visual place recognition (VPR) aims to determine the general geographical location of a query image by retrieving visually similar images from a large geo-tagged database. To obtain a global representation for each place image, most approaches typically focus on the aggregation of deep features extracted from a backbone through using current prominent architectures (e.g., CNNs, MLPs, pooling layer, and transformer encoder), giving little attention to the transformer decoder. However, we argue that its strong capability to capture contextual dependencies and generate accurate features holds considerable potential for the VPR task. To this end, we propose an Efficient Decoder Transformer (EDTformer) for feature aggregation, which consists of several stacked simplified decoder blocks followed by two linear layers to directly produce robust and discriminative global representations. Specifically, we do this by formulating deep features as the keys and values, as well as a set of learnable parameters as the queries. Our EDTformer can fully utilize the contextual information within deep features, then gradually decode and aggregate the effective features into the learnable queries to output the global representations. Moreover, to provide more powerful deep features for EDTformer and further facilitate the robustness, we use the foundation model DINOv2 as the backbone and propose a Low-rank Parallel Adaptation (LoPA) method to enhance its performance in VPR, which can refine the intermediate features of the backbone progressively in a memory- and parameter-efficient way. As a result, our method not only outperforms single-stage VPR methods on multiple benchmark datasets, but also outperforms two-stage VPR methods which add a re-ranking with considerable cost. Code will be available at https://github.com/Tong-Jin01/EDTformer.

cs.CV↗

arXiv2025

A Hybrid Cloud Management Plane for Data Processing Pipelines

As organizations increasingly rely on data-driven insights, the ability to run data intensive applications seamlessly across multiple cloud environments becomes critical for tapping into cloud innovations while complying with various security and regulatory requirements. However, big data application development and deployment remain challenging to accomplish in such environments. With the increasing containerization and modernization of big data applications, we argue that a unified control/management plane now makes sense for running these applications in hybrid cloud environments. To this end, we study the problem of building a generic hybrid-cloud management plane to radically simplify managing big data applications. A generic architecture for hybrid-cloud management, called Titchener, is proposed in this paper. Titchener comprises of independent and loosely coupled local control planes interacting with a highly available public cloud hosted global management plane. We describe a possible instantiation of Titchener based on Kubernetes and address issues related to global service discovery, network connectivity and access control enforcement. We also validate our proposed designs with a real management plane implementation based on a popular big data workflow orchestration in hybrid-cloud environments.

cs.DC↗

arXiv2025

GazeSwipe: Enhancing Mobile Touchscreen Reachability through Seamless Gaze and Finger-Swipe Integration

Smartphones with large screens provide users with increased display and interaction space but pose challenges in reaching certain areas with the thumb when using the device with one hand. To address this, we introduce GazeSwipe, a multimodal interaction technique that combines eye gaze with finger-swipe gestures, enabling intuitive and low-friction reach on mobile touchscreens. Specifically, we design a gaze estimation method that eliminates the need for explicit gaze calibration. Our approach also avoids the use of additional eye-tracking hardware by leveraging the smartphone's built-in front-facing camera. Considering the potential decrease in gaze accuracy without dedicated eye trackers, we use finger-swipe gestures to compensate for any inaccuracies in gaze estimation. Additionally, we introduce a user-unaware auto-calibration method that improves gaze accuracy during interaction. Through extensive experiments on smartphones and tablets, we compare our technique with various methods for touchscreen reachability and evaluate the performance of our auto-calibration strategy. The results demonstrate that our method achieves high success rates and is preferred by users. The findings also validate the effectiveness of the auto-calibration strategy.

cs.HC↗

arXiv2025

3D Prior is All You Need: Cross-Task Few-shot 2D Gaze Estimation

3D and 2D gaze estimation share the fundamental objective of capturing eye movements but are traditionally treated as two distinct research domains. In this paper, we introduce a novel cross-task few-shot 2D gaze estimation approach, aiming to adapt a pre-trained 3D gaze estimation network for 2D gaze prediction on unseen devices using only a few training images. This task is highly challenging due to the domain gap between 3D and 2D gaze, unknown screen poses, and limited training data. To address these challenges, we propose a novel framework that bridges the gap between 3D and 2D gaze. Our framework contains a physics-based differentiable projection module with learnable parameters to model screen poses and project 3D gaze into 2D gaze. The framework is fully differentiable and can integrate into existing 3D gaze networks without modifying their original architecture. Additionally, we introduce a dynamic pseudo-labelling strategy for flipped images, which is particularly challenging for 2D labels due to unknown screen poses. To overcome this, we reverse the projection process by converting 2D labels to 3D space, where flipping is performed. Notably, this 3D space is not aligned with the camera coordinate system, so we learn a dynamic transformation matrix to compensate for this misalignment. We evaluate our method on MPIIGaze, EVE, and GazeCapture datasets, collected respectively on laptops, desktop computers, and mobile devices. The superior performance highlights the effectiveness of our approach, and demonstrates its strong potential for real-world applications.

cs.CV↗