Enhanced Semantic Alignment in Transformer Tracking via Position Learning and Force-Directed Attention [0.03%]
基于位置学习和力导向注意力的变压器跟踪中增强的语义对齐方法
Mingyu Cao,Haotian Wang,Haoxuan Li et al.
Mingyu Cao et al.
In the field of visual object tracking, one-stream pipelines have become the mainstream framework due to their efficient integration of feature extraction and relationship modeling. However, existing methods still face the issue of semantic...
Anticipating Object Interactions Via Aggregation and Distillation of Spatio-Temporal Knowledge From Vision Language Models [0.03%]
通过视觉语言模型的空间 temporal知识聚合和提炼预测物体交互
Yang Liu,Dejie Yang,Minghang Zheng et al.
Yang Liu et al.
The increasing prevalence of wearable cameras has driven the development of egocentric (first-person) systems that assist human activities proactively by anticipating imminent interactions. A central challenge in this domain is active objec...
SPD Matrix Learning for Neuroimaging Analysis: Perspectives, Methods, and Challenges [0.03%]
基于SPD矩阵的学习在神经影像分析中的应用：现状、方法及挑战
Ce Ju,Reinmar Kobler,Antoine Collas et al.
Ce Ju et al.
Neuroimaging provides essential tools for characterizing brain activity and inter-regional connectivity through modalities that capture complementary aspects of brain organization. At the same time, extracting meaningful neural signatures r...
Exploiting Vision Language Model for Training-Free 3D Point Cloud Understanding via Improved Graph Score Propagation [0.03%]
基于改进图分数传播的无需训练的视觉语言模型点云理解方法
Tiankai Chen,Yushu Li,Adam Goodge et al.
Tiankai Chen et al.
3D point cloud understanding, e.g. out-of-distribution detection and recognition, remains a challenge, particularly in applications where safe and robust perception is critical. While recent 3D VLMs enable direct recognition via text prompt...
Wenfeng Song,Xingliang Jin,Shuai Li et al.
Wenfeng Song et al.
Multi-modality motion stylization presents a solution to the challenge of generating flexible, stylized motion based on multimodal style inputs. Historically, motion stylization has grappled with the difficulty of balancing content and styl...
Tianyu Zhu,Ying Fu,Hesong Li et al.
Tianyu Zhu et al.
RAW video restoration is fundamental to high-quality low-level perception and serves as the basis for a wide range of downstream vision applications. While binary neural networks (BNNs) enable efficient lightweight deployment for image enha...
Exposition on Over-Squashing Problem of GNNs: Current Methods, Benchmarks and Challenges [0.03%]
GNN过平滑问题详解：当前方法、基准测试与挑战
Dai Shi,Andi Han,Lequan Lin et al.
Dai Shi et al.
Graph-based message-passing neural networks (MPNNs) have achieved remarkable success in both node and graph-level tasks. However, several identified problems, including over-smoothing (OSM), limited expressive power, and over-squashing (OSQ...
Training-free Context-adaptive Attention for Efficient Long Context Modeling [0.03%]
无需训练的上下文自适应注意力，用于高效的长上下文建模
Zeng You,Yaofo Chen,Shuhai Zhang et al.
Zeng You et al.
Large Language Models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing tasks. These capabilities stem primarily from the self-attention mechanism, which enables the modeling of long-range d...
Xingwei Zhang,Gang Zhou,Xiaolong Zheng et al.
Xingwei Zhang et al.
Deep hashing has emerged as an effective and widely adopted framework for similarity retrieval, owing to its computational efficiency and the powerful feature extraction capabilities of deep learning (DL). Despite their strong performance i...
Towards Heterogeneous-Degradation-Robust Image Fusion with Controllable Generative Modulation [0.03%]
一种可控的生成式图像融合模块化网络用于异质退化图像恢复
Lei Cao,Hao Zhang,Peng Zhang et al.
Lei Cao et al.
Enhancing degradation robustness is essential for deploying image fusion techniques in real-world dynamic scenes. However, most existing methods either handle a single degradation type or assume fixed multi-degradation settings, making them...