11 papers
On-Policy Distillation for Vision-Language Model Adaptation, an Effective Paradigm on Low-Quality Multimodal Data
Hongyuan Zhang, Xianda Guo, Yanlun Peng +6
Knowledge distillation offers an efficient route to transfer a task-adapted vision-language teacher to a compact student. The training target in current vision-language distillatio…
Semantic-Aware Temporal Adaptation for UAV Anti-UAV Tracking
Xiaozhen Qiao, Da Zhang, Yubin Guo +3
UAV Anti-UAV tracking is an emerging low-altitude security task for localizing an adversarial UAV using the onboard camera of a moving observer UAV. It differs from conventional UA…
AHAP: Reconstructing Arbitrary Humans from Arbitrary Perspectives with Geometric Priors
Xiaozhen Qiao, Wenjia Wang, Zhiyuan Zhao +4
Reconstructing 3D humans from images captured at multiple perspectives typically requires pre-calibration, like using checkerboards or MVS algorithms, which limits scalability and…
EgoGrasp: World-Space Hand-Object Interaction Estimation from Egocentric Videos
Hongming Fu, Wenjia Wang, Xiaozhen Qiao +5
We propose EgoGrasp, the first method to reconstruct world-space hand-object interactions (W-HOI) from dynamic egoview videos, supporting open-vocabulary objects. Accurate W-HOI re…
DeCo-VAE: Learning Compact Latents for Video Reconstruction via Decoupled Representation
Xiangchen Yin, Jiahui Yuan, Zhangchi Hu +5
Existing video Variational Autoencoders (VAEs) generally overlook the similarity between frame contents, leading to redundant latent modeling. In this paper, we propose decoupled V…
Mitigating Long-Tail Bias in HOI Detection via Adaptive Diversity Cache
Yuqiu Jiang, Xiaozhen Qiao, Yifan Chen +3
Human-Object Interaction (HOI) detection is a fundamental task in computer vision, empowering machines to comprehend human-object relationships in diverse real-world scenarios. Rec…
Class-Aware Prototype Learning with Negative Contrast for Test-Time Adaptation of Vision-Language Models
Xiaozhen Qiao, Jingkai Zhao, Yuqiu Jiang +4
Vision-Language Models (VLMs) demonstrate impressive zero-shot generalization through large-scale image-text pretraining, yet their performance can drop once the deployment distrib…
Bridging Perception and Planning: Towards End-to-End Planning for Signal Temporal Logic Tasks
Bowen Ye, Junyue Huang, Yang Liu +2
We investigate the task and motion planning problem for Signal Temporal Logic (STL) specifications in robotics. Existing STL methods rely on pre-defined maps or mobility representa…
Bidirectional Prototype-Reward co-Evolution for Test-Time Adaptation of Vision-Language Models
Xiaozhen Qiao, Peng Huang, Jiakang Yuan +6
Test-time adaptation (TTA) is crucial in maintaining performance of Vision Language Models (VLMs) when facing distribution shifts, particularly when the source data or target label…
ComKD-CLIP: Comprehensive Knowledge Distillation for Contrastive Language-Image Pre-traning Model
Yifan Chen, Xiaozhen Qiao, Zhe Sun +1
Contrastive Language-Image Pre-training (CLIP) models excel in integrating semantic information between images and text through contrastive learning techniques. It has achieved rem…
CREST: Cross-modal Resonance through Evidential Deep Learning for Enhanced Zero-Shot Learning
Haojian Huang, Xiaozhen Qiao, Zhuo Chen +5
Zero-shot learning (ZSL) enables the recognition of novel classes by leveraging semantic knowledge transfer from known to unknown categories. This knowledge, typically encapsulated…