Xianghui Pan

Dynamics Are Learned, Not Told: Semi-Supervised Discovery of Latent Dynamics Geometries For Zero-Shot Policy Adaptation

Zhiming Xu, Weitao Zhou, Xianghui Pan, Nanshan Deng et autres

Real-world dynamics shifts pose a critical challenge for reinforcement learning in robotics, as policies tightly coupled to nominal environments often fail catastrophically when physical conditions change. Most existing methods rely on encoding explicitly identified physical parameters into a latent context, a parameter-centric …

Dynamics Are Learned, Not Told: Semi-Supervised Discovery of Latent Dynamics Geometries For Zero-Shot Policy Adaptation

Zhiming Xu, Weitao Zhou, Xianghui Pan, Nanshan Deng et autres

Real-world dynamics shifts pose a critical challenge for reinforcement learning in robotics, as policies tightly coupled to nominal environments often fail catastrophically when physical conditions change. Most existing methods rely on encoding explicitly identified physical parameters into a latent context, a parameter-centric …

cn, us (code pays fourni par la source)

Deconfounded Lifelong Learning for Autonomous Driving via Dynamic Knowledge Spaces

Jiayuan Du, Yuebing Song, Yiming Zhao, Xianghui Pan et autres

End-to-End autonomous driving (E2E-AD) systems face challenges in lifelong learning, including catastrophic forgetting, difficulty in knowledge transfer across diverse scenarios, and spurious correlations between unobservable confounders and true driving intents. To address these issues, we propose DeLL, a Deconfounded Lifelong Learning framework …

cn (code pays fourni par la source)

BEVDrive-E2E: Imitation With Bird's Eye View Perception for Interpretable End-to-End Autonomous Driving

Jiayuan Du, Yuebing Song, Xianghui Pan, Shuai Su et autres

Imitation learning (IL) for end-to-end autonomous driving (E2E-AD) has made great progress recently in the closed-loop evaluation of the CARLA simulator. However, the causal confusion remains an open problem. To address this issue, we propose the BEVDrive-E2E to explore the interpretability of …

cn (code pays fourni par la source)

Deconfounded Lifelong Learning for Autonomous Driving via Dynamic Knowledge Spaces

Jiayuan Du, Yuebing Song, Yiming Zhao, Xianghui Pan et autres

cn (code pays fourni par la source)

MaRIne: An Underwater Monocular Real-Time Object Detection and Geometric Estimation Framework Based on Motion Inversion

Yuebing Song, Xianghui Pan, Junji Zhu, Jiayuan Du et autres

cn (code pays fourni par la source)

Rotation-Equivariant Robot Vision: A Perspective via Correspondence-Matching and Pre-training

Shuai Su, Xianghui Pan, Jiayuan Du, Chengju Liu et autres

Correspondence matching is a fundamental and crucial task in robot vision. In recent years, deep learning-based keypoint matching techniques have shown outstanding performance in downstream tasks. Conventional learning-based correspondence matching methods rely on large datasets and a specific training procedure. Correspondence techniques …

cn (code pays fourni par la source)

LGPR: Local Feature Learning Brings More Generalizable Visual Place Recognition

Shuai Su, Jingwei Yang, Jiayuan Du, Xianghui Pan et autres

We propose a Visual Place Recognition (VPR) framework by sharing lightweight keypoint extraction modules for local features. Current research on the joint learning of local keypoint matching and VPR is relatively scarce, and the application deployment of real-time spatial computing on edge …

cn (code pays fourni par la source)

Polyanhydride‐Based Microparticles for Programmable Pulsatile Release of Diphtheria Toxoid (DT) for Single‐Injection Self‐Boosting Vaccines

Linzixuan Zhang, Rui-Qing Xiao, Wenhao Gao, Johnny Garcia et autres

Vaccination remains a critical tool in preventing infectious diseases, yet its effectiveness is undermined by under-immunization, particularly for vaccines requiring multiple doses that patients fail to complete. To address this challenge, the development of single-injection platforms delivering self-boosting vaccines has gained significant …

us, kr (code pays fourni par la source)

Degradable poly(β-amino ester) microparticles for cleansing products and food fortification

Linzixuan Zhang, Rui-Qing Xiao, Tianyi Jin, Xianghui Pan et autres

Microplastic pollution is a pressing global crisis caused by the extensive use of nondegradable microplastic materials in daily activities. One effective approach to mitigate this issue is to replace nondegradable plastics with degradable materials that have properties amendable for targeted applications. Here …

us (code pays fourni par la source)

GenerOcc: Self-supervised Framework of Real-time 3D Occupancy Prediction for Monocular Generic Cameras

Xianghui Pan, Jiayuan Du, Shuai Su, Wenhao Zong et autres

In the context of 3D scene perception tasks, the significance of 3D occupancy prediction has been progressively growing, aiming to forecast the occupancy state of voxels in a discrete 3D space. However, existing methods typically exhibit several limitations, such as restricted adaptability …

cn (code pays fourni par la source)

DVT: Decoupled Dual-Branch View Transformation for Monocular Bird’s Eye View Semantic Segmentation

Jiayuan Du, Xianghui Pan, Mengjiao Shen, Shuai Su et autres

Monocular Bird’s Eye View (BEV) semantic segmentation is critical for autonomous driving for its inherent advantages in spatial representation and downstream tasks. However, it is challenging to simultaneously learn view transformation and pixel-wise classification. Previous works suffer from non-flat region distortion, distant …

cn (code pays fourni par la source)