6 papers · 1 filter
GaussianSeed: Hierarchical Gaussian Seeding for High-Resolution 3D Occupancy Prediction
Xinzhuo Li, Xianghui Pan, Jiayuan Du +4
Vision-centric 3D occupancy prediction provides dense scene representations essential for autonomous driving and robotic navigation, yet existing methods struggle to scale to high…
MemPose: Category-level Object Pose Estimation with Memory
Xiao Lin, Minghao Zhu, Yun Peng +4
In the pursuit of robust and generalizable category-level object pose estimation, most existing methods adopt parametric formulations that learn effective representations from data…
SAM-LAD: Segment Anything Model Meets Zero-Shot Logic Anomaly Detection
Yun Peng, Xiao Lin, Nachuan Ma +4
Visual anomaly detection is vital in real-world applications, such as industrial defect detection and medical diagnosis. However, most existing methods focus on local structural an…
Efficient Text-driven Motion Generation via Latent Consistency Training
Mengxian Hu, Minghao Zhu, Xun Zhou +4
Text-driven human motion generation based on diffusion strategies establishes a reliable foundation for multimodal applications in human-computer interactions. However, existing ad…
Fine-Grained Spatiotemporal Motion Alignment for Contrastive Video Representation Learning
Minghao Zhu, Xiao Lin, Ronghao Dang +2
As the most essential property in a video, motion information is critical to a robust and generalized video representation. To inject motion dynamics, recent works have adopted fra…
MoTE: Reconciling Generalization with Specialization for Visual-Language to Video Knowledge Transfer
Minghao Zhu, Zhengpu Wang, Mengxian Hu +5
Transferring visual-language knowledge from large-scale foundation models for video recognition has proved to be effective. To bridge the domain gap, additional parametric modules…