TAT-VPR: Ternary Adaptive Transformer for Dynamic and Efficient Visual Place Recognition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Grainge, Oliver, Milford, Michael, Bodala, Indu, Ramchurn, Sarvapali D., Ehsan, Shoaib
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908374336012288
author Grainge, Oliver
Milford, Michael
Bodala, Indu
Ramchurn, Sarvapali D.
Ehsan, Shoaib
author_facet Grainge, Oliver
Milford, Michael
Bodala, Indu
Ramchurn, Sarvapali D.
Ehsan, Shoaib
contents TAT-VPR is a ternary-quantized transformer that brings dynamic accuracy-efficiency trade-offs to visual SLAM loop-closure. By fusing ternary weights with a learned activation-sparsity gate, the model can control computation by up to 40% at run-time without degrading performance (Recall@1). The proposed two-stage distillation pipeline preserves descriptor quality, letting it run on micro-UAV and embedded SLAM stacks while matching state-of-the-art localization accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2505_16447
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TAT-VPR: Ternary Adaptive Transformer for Dynamic and Efficient Visual Place Recognition
Grainge, Oliver
Milford, Michael
Bodala, Indu
Ramchurn, Sarvapali D.
Ehsan, Shoaib
Computer Vision and Pattern Recognition
TAT-VPR is a ternary-quantized transformer that brings dynamic accuracy-efficiency trade-offs to visual SLAM loop-closure. By fusing ternary weights with a learned activation-sparsity gate, the model can control computation by up to 40% at run-time without degrading performance (Recall@1). The proposed two-stage distillation pipeline preserves descriptor quality, letting it run on micro-UAV and embedded SLAM stacks while matching state-of-the-art localization accuracy.
title TAT-VPR: Ternary Adaptive Transformer for Dynamic and Efficient Visual Place Recognition
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.16447