MGGA: Make GeM Great Again via Regularization Branch to Mitigate Channel Vanishing in Visual Place Recognition
學年 114
學期 2
發表日期 2026-04-10
作品名稱 MGGA: Make GeM Great Again via Regularization Branch to Mitigate Channel Vanishing in Visual Place Recognition
作品名稱（其他語言）
著者 Qixi Zhao; Jiwei Nie; Zuotao Ning; Joe-Mei Feng
作品所屬單位
出版者
會議名稱 CVM 2026
會議地點 Seoul, South Korea
摘要 Deep-learning-based methods have achieved significant success in the Visual Place Recognition (VPR) task, which is important for autonomous driving and robotics systems. Recent advancements primarily focus on the sophisticated feature aggregation module. This paper argues for a shift in emphasis toward the backbone features. Through an in-depth analysis of GeM, one of the simplest pooling aggregator based VPR method, we identify a prevalent issue, termed ’Channel vanishing’. The issue manifests as a substantial proportion of channels in both the final GeM descriptor and the backbone output local features turning zero-valued and inactive during training, thereby drastically diminishing the representational capacity of the model and undermining its VPR performance. In order to solve this problem, we propose a regularization branch with a fully connected layer for the GeM pipeline. This branch successfully mitigates Channel vanishing and further enriches the diversity and representation of the backbone output features. During inference, our streamlined model, using only the GeM aggregator, achieves state-of-the-art performance among backbones that are not transformerbased. Notably, when utilizing the DINOv2-B backbone, our method derives 99.1% recall@1 and 100% recall@5 VPR scores on the Tokyo24/7 dataset. This result suggests that strengthening backbone features can substantially narrow the gap between simple GeM pooling and more complex aggregators; assessing how broadly this observation transfers to other aggregators is an interesting direction.
關鍵字 Deep-learning; Visual Place Recognition; Autonomous Navigation; Robotics; GeM; Channel vanishing
語言 zh_TW
收錄於
會議性質 國內
校內研討會地點 無
研討會時間 20260410~20260412
通訊作者
國別 TWN
公開徵稿
出版型式
出處
相關連結

機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/129203 )