arXiv subjects

Zhaoyang Lu

Publications and source records attributed to Zhaoyang Lu.

7 recordsLinked to original sources

arXiv2026

GLAM: Training a latent world model over global spatiotemporal memory for active exploration and navigation

Active exploration and semantic navigation require an embodied agent to build memory from partial observations, predict how the evolution of observed spatial memory may support future motion, and convert that prediction into actionable plans. We present GLAM, a goal-conditioned latent world model trained over global spatiotemporal memory, and GLAM NAV, the complete navigation system built around it. Given historical map tokens, a navigation goal, and the current robot pose, GLAM jointly predicts future map representations and robot-centric waypoint latents, allowing future spatial context and navigation intent to be inferred in a shared representation space. The model follows a JEPA-like latent prediction paradigm, operates directly on map-level latent tokens rather than RGB reconstruction, and uses a pretrained waypoint encoder-decoder to supervise and decode navigation plans within GLAM NAV. Training data are collected by replaying ObjectNav expert trajectories in Habitat over HM3D v0.2 scene assets and slicing them into multi-timescale prediction samples. On a controlled HM3D-ObjectNav subset reproduction setting, GLAM NAV improves over a reproduced BSC-Nav baseline in both success rate and success weighted by path length.

cs.RO↗

arXiv2025

Impact of fast ions on turbulent transport in high-\b{eta} HL-2A scenarios

The fast-ion (FI) on turbulent transport is one of the key topics of magnetic confinement fusion. This work focus on the impact of FI pressure gradients on turbulence in a high-\b{eta} plasma scenario using gyrokinetic simulations. Linear analyses reveal that FIs strongly stabilize ion temperature gradient (ITG) modes via the thermal-ion dilution, while their influence on trapped electron modes (TEMs) is minimal. At elevated FI pressure gradients, a transition to a FI-driven BAE (FI-BAE) regime occurs, as evidenced by mode structure and frequency alignment within the Alfvénic gap. Electron \b{eta} scans further demonstrate the emergence of kinetic ballooning modes (KBMs) at higher \b{eta}, whereas an ITG-TEM hybrid turbulence dominates near experimental \b{eta} values. Nonlinear simulations show that moderate FI pressure suppresses transport via zonal flow (ZF) shear, whereas strong FI drive weakens ZFs and enhances transport by destabilizing FI-BAEs. These results highlight the dual role of FIs in regulating turbulence and offer insight into multiscale transport physics relevant for high-performance plasmas.

physics.plasm-ph↗

arXiv2023

Long radial coherence of electron temperature fluctuations in non-local transport in HL-2A plasmas

The dynamics of long-wavelength ($k_θ<1.4 \mathrm{\ cm^{-1}}$), broadband (20-200 kHz) electron temperature fluctuations ($\tilde T_e/T_e$) of plasmas in gas-puff experiments were observed for the first time in HL-2A tokamak. In a relative low density ($n_e(0) \simeq 0.91 \sim 1.20 \times10^{19}/m^3$) scenario, after gas-puffing the core temperature increases and the edge temperature drops. On the contrary, temperature fluctuation drops at the core and increases at the edge. Analyses show the non-local emergence is accompanied with a long radial coherent length of turbulent fluctuations. While in a higher density ($n_e(0) \simeq 1.83 \sim 2.02 \times10^{19}/m^3$) scenario, the phenomena were not observed. Furthermore, compelling evidence indicates that $\textbf{E} \times \textbf{B}$ shear serves as a substantial contributor to this extensive radial interaction. This finding offers a direct explanatory link to the intriguing core-heating phenomenon witnessed within the realm of non-local transport.

physics.plasm-ph↗

arXiv2019

Deep Image-to-Video Adaptation and Fusion Networks for Action Recognition

Existing deep learning methods for action recognition in videos require a large number of labeled videos for training, which is labor-intensive and time-consuming. For the same action, the knowledge learned from different media types, e.g., videos and images, may be related and complementary. However, due to the domain shifts and heterogeneous feature representations between videos and images, the performance of classifiers trained on images may be dramatically degraded when directly deployed to videos. In this paper, we propose a novel method, named Deep Image-to-Video Adaptation and Fusion Networks (DIVAFN), to enhance action recognition in videos by transferring knowledge from images using video keyframes as a bridge. The DIVAFN is a unified deep learning model, which integrates domain-invariant representations learning and cross-modal feature fusion into a unified optimization framework. Specifically, we design an efficient cross-modal similarities metric to reduce the modality shift among images, keyframes and videos. Then, we adopt an autoencoder architecture, whose hidden layer is constrained to be the semantic representations of the action class names. In this way, when the autoencoder is adopted to project the learned features from different domains to the same space, more compact, informative and discriminative representations can be obtained. Finally, the concatenation of the learned semantic feature representations from these three autoencoders are used to train the classifier for action recognition in videos. Comprehensive experiments on four real-world datasets show that our method outperforms some state-of-the-art domain adaptation and action recognition methods.

cs.CV↗

arXiv2019

Global Temporal Representation based CNNs for Infrared Action Recognition

Infrared human action recognition has many advantages, i.e., it is insensitive to illumination change, appearance variability, and shadows. Existing methods for infrared action recognition are either based on spatial or local temporal information, however, the global temporal information, which can better describe the movements of body parts across the whole video, is not considered. In this letter, we propose a novel global temporal representation named optical-flow stacked difference image (OFSDI) and extract robust and discriminative feature from the infrared action data by considering the local, global, and spatial temporal information together. Due to the small size of the infrared action dataset, we first apply convolutional neural networks on local, spatial, and global temporal stream respectively to obtain efficient convolutional feature maps from the raw data rather than train a classifier directly. Then these convolutional feature maps are aggregated into effective descriptors named three-stream trajectory-pooled deep-convolutional descriptors by trajectory-constrained pooling. Furthermore, we improve the robustness of these features by using the locality-constrained linear coding (LLC) method. With these features, a linear support vector machine (SVM) is adopted to classify the action data in our scheme. We conduct the experiments on infrared action recognition datasets InfAR and NTU RGB+D. The experimental results show that the proposed approach outperforms the representative state-of-the-art handcrafted features and deep learning features based methods for the infrared action recognition.

cs.CV↗

arXiv2019

Transferable Feature Representation for Visible-to-Infrared Cross-Dataset Human Action Recognition

Recently, infrared human action recognition has attracted increasing attention for it has many advantages over visible light, that is, being robust to illumination change and shadows. However, the infrared action data is limited until now, which degrades the performance of infrared action recognition. Motivated by the idea of transfer learning, an infrared human action recognition framework using auxiliary data from visible light is proposed to solve the problem of limited infrared action data. In the proposed framework, we first construct a novel Cross-Dataset Feature Alignment and Generalization (CDFAG) framework to map the infrared data and visible light data into a common feature space, where Kernel Manifold Alignment (KEMA) and a dual alignedto-generalized encoders (AGE) model are employed to represent the feature. Then, a support vector machine (SVM) is trained, using both the infrared data and visible light data, and can classify the features derived from infrared data. The proposed method is evaluated on InfAR, which is a publicly available infrared human action dataset. To build up auxiliary data, we set up a novel visible light action dataset XD145. Experimental results show that the proposed method can achieve state-of-the-art performance compared with several transfer learning and domain adaptation methods.

cs.CV↗

arXiv2019

Hierarchically Learned View-Invariant Representations for Cross-View Action Recognition

Recognizing human actions from varied views is challenging due to huge appearance variations in different views. The key to this problem is to learn discriminant view-invariant representations generalizing well across views. In this paper, we address this problem by learning view-invariant representations hierarchically using a novel method, referred to as Joint Sparse Representation and Distribution Adaptation (JSRDA). To obtain robust and informative feature representations, we first incorporate a sample-affinity matrix into the marginalized stacked denoising Autoencoder (mSDA) to obtain shared features, which are then combined with the private features. In order to make the feature representations of videos across views transferable, we then learn a transferable dictionary pair simultaneously from pairs of videos taken at different views to encourage each action video across views to have the same sparse representation. However, the distribution difference across views still exists because a unified subspace where the sparse representations of one action across views are the same may not exist when the view difference is large. Therefore, we propose a novel unsupervised distribution adaptation method that learns a set of projections that project the source and target views data into respective low-dimensional subspaces where the marginal and conditional distribution differences are reduced simultaneously. Therefore, the finally learned feature representation is view-invariant and robust for substantial distribution difference across views even the view difference is large. Experimental results on four multiview datasets show that our approach outperforms the state-ofthe-art approaches.

cs.CV↗