.png){kind=icon}
返回
One-Stream Vision-Language Memory Network for Object Tracking
.png){kind=icon}
PRE
AI
DOI：10.1109/TMM.2023.3285441.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Most existing tracking methods try to represent the target by exploiting visual information as much as possible based on the various deep networks. However, the appearance model hardly describes the attribute feature of the target well, which makes the trackers fail to adapt to the complex visual surrounding. In this article, inspired by brain-like intelligence, we propose an One-stream Vision-Language Memory network (OVLM) for object tracking. Firstly, we use the combination of vision and language to build the target model and use the semantic information in the language to compensate for the instability of visual information, making the target model more stable in the face of complex appearance changes. Secondly, to build a more compact target model, we propose a memory token selection mechanism that utilizes linguistic information to eliminate tokens that do not contain target information. Furthermore, to provide better visual information for target modeling, we propose a language-based evaluation method to select high-quality target samples to be stored in the memory. Finally, OVLM achieves a 64.7% success rate on the large-scale tracking benchmark dataset TNL2K, outperforming the previous best result (VLT) by 11.6%. By exposing the possibility of the vision-language memory network, we aim to draw greater attention to it and open up new avenues for vision-language tracking.
Keyword:
Target tracking
Visualization
Linguistics
Iron
Feature extraction
Computational modeling
Adaptation models
Object tracking
vision-language
one-stream
memory network
期刊
IF:
9.7
论文数:
4.5K
被引数:
2.4W
机构
引用论文
Ultrasound evaluation of hand, wrist and foot joint synovitis in systemic lupus erythematosus.png)
Rheumatology
OAAI
RheumatologyIF0
OAAI分享收藏Cobalt oxides nanoparticles supported on nitrogen-doped carbon nanotubes as high-efficiency cathode catalysts for microbial fuel cells负载在氮掺杂碳纳米管上的钴氧化物纳米颗粒作为微生物燃料电池的高效阴极催化剂OAAI
OAAI分享收藏A novel flake-ball-like magnetic Fe3O4/γ-MnO2 meso-porous nano-composite: Adsorption of fluorinion and effect of water chemistryChemosphere
ChemosphereIF0
PREAI
分享收藏
查看更多内容