activity
20242026
collaborators

7 papers

cs.CV2026

Towards Geometry-Aware and Motion-Guided Video Human Mesh Recovery

Hongjun Chen, Huan Zheng, Wencheng Han +1

Existing video-based 3D Human Mesh Recovery (HMR) methods often produce physically implausible results, stemming from their reliance on flawed intermediate 3D pose anchors and thei…

cs.CV2026

From Human Intention to Action Prediction: Intention-Driven End-to-End Autonomous Driving

Huan Zheng, Yucheng Zhou, Tianyi Yan +9

While end-to-end autonomous driving has achieved remarkable progress in geometric control, current systems remain constrained by a command-following paradigm that relies on simple…

cs.CV2025

Rethinking the Encoding and Annotating of 3D Bounding Box: Corner-Aware 3D Object Detection from Point Clouds

Qinghao Meng, Junbo Yin, Jianbing Shen +1

Center-aligned regression remains dominant in LiDAR-based 3D object detection, yet it suffers from fundamental instability: object centers often fall in sparse or empty regions of…

cs.CV2025

Semantic Causality-Aware Vision-Based 3D Occupancy Prediction

Dubing Chen, Huan Zheng, Yucheng Zhou +5

Vision-based 3D semantic occupancy prediction is a critical task in 3D vision that integrates volumetric 3D reconstruction with semantic understanding. Existing methods, however, o…

cs.CV2025

Rethinking Temporal Fusion with a Unified Gradient Descent View for 3D Semantic Occupancy Prediction

Dubing Chen, Huan Zheng, Jin Fang +6

We present GDFusion, a temporal fusion method for vision-based 3D semantic occupancy prediction (VisionOcc). GDFusion opens up the underexplored aspects of temporal fusion within t…

cs.CV2025

Breaking Down Monocular Ambiguity: Exploiting Temporal Evolution for 3D Lane Detection

Huan Zheng, Wencheng Han, Tianyi Yan +2

Monocular 3D lane detection aims to estimate the 3D position of lanes from frontal-view (FV) images. However, existing methods are fundamentally constrained by the inherent ambigui…