Why-so-deep: Towards Boosting Previously Trained Models For Visual Place Recognition
2022 · M. Usman Maqbool Bhutta, Yuxiang Sun, Darwin Lau, et al.
Abstract
Deep learning-based image retrieval techniques for the loop closure detection demonstrate satisfactory performance. However, it is still challenging to achieve high-level performance based on previously trained models in different geographical regions. This paper addresses the problem of their deployment with simultaneous localization and mapping (SLAM) systems in the new environment. The general baseline approach uses additional information, such as GPS, sequential keyframes tracking, and re-training the whole environment to enhance the recall rate. We propose a novel approach for improving image retrieval based on previously trained models. We present an intelligent method, MAQBOOL, to amplify the power of pre-trained models for better image recall and its application to real-time multiagent SLAM systems. We achieve comparable image retrieval results at a low descriptor dimension (512-D), compared to the high descriptor dimension (4096-D) of state-of-the-art methods. We use spatial i
Authors
(none)
Tags
Stats
Related papers
- DALG: Deep Attentive Local And Global Modeling For Image Retrieval (2022)0.00
- Boq: A Place Is Worth A Bag Of Learnable Queries (2024)16.09
- End-to-end Learning Of Deep Visual Representations For Image Retrieval (2016)19.66
- Multires-netvlad: Augmenting Place Recognition Training With Low-resolution Imagery (2022)16.01
- Deep Image Retrieval: Learning Global Representations For Image Search (2016)19.67
- Are Local Features All You Need For Cross-domain Visual Place Recognition? (2023)13.80
- Probabilistic Visual Place Recognition For Hierarchical Localization (2021)11.08
- Megaloc: One Retrieval To Place Them All (2025)9.19