Semantic-focused Patch Tokenizer with Multi-branch Mixer for Visual Place Recognition
Zhenyu Xu,Ziliang Ren,Qieshi Zhang,Jie Lou,Dacheng Tao,Jun Cheng,Zhenyu Xu,Ziliang Ren,Qieshi Zhang,Jie Lou,Dacheng Tao,Jun Cheng
Visual Place Recognition (VPR) is critical for navigation and loop closure in autonomous driving tasks, mitigating the impact of shift errors caused by dynamic changes in the environment. Due to the limited ability of backbone networks and extreme environmental changes, current methods fail to capture foundational semantic details that include the distinctive attributes for unique place identifica...
