Eff-GRot: Efficient and Generalizable Rotation Estimation with Transformers

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mathioulakis, Fanis, Radevski, Gorjan, Tuytelaars, Tinne
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918257867358208
author Mathioulakis, Fanis
Radevski, Gorjan
Tuytelaars, Tinne
author_facet Mathioulakis, Fanis
Radevski, Gorjan
Tuytelaars, Tinne
contents We introduce Eff-GRot, an approach for efficient and generalizable rotation estimation from RGB images. Given a query image and a set of reference images with known orientations, our method directly predicts the object's rotation in a single forward pass, without requiring object- or category-specific training. At the core of our framework is a transformer that performs a comparison in the latent space, jointly processing rotation-aware representations from multiple references alongside a query. This design enables a favorable balance between accuracy and computational efficiency while remaining simple, scalable, and fully end-to-end. Experimental results show that Eff-GRot offers a promising direction toward more efficient rotation estimation, particularly in latency-sensitive applications.
format Preprint
id arxiv_https___arxiv_org_abs_2512_18784
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Eff-GRot: Efficient and Generalizable Rotation Estimation with Transformers
Mathioulakis, Fanis
Radevski, Gorjan
Tuytelaars, Tinne
Computer Vision and Pattern Recognition
Machine Learning
We introduce Eff-GRot, an approach for efficient and generalizable rotation estimation from RGB images. Given a query image and a set of reference images with known orientations, our method directly predicts the object's rotation in a single forward pass, without requiring object- or category-specific training. At the core of our framework is a transformer that performs a comparison in the latent space, jointly processing rotation-aware representations from multiple references alongside a query. This design enables a favorable balance between accuracy and computational efficiency while remaining simple, scalable, and fully end-to-end. Experimental results show that Eff-GRot offers a promising direction toward more efficient rotation estimation, particularly in latency-sensitive applications.
title Eff-GRot: Efficient and Generalizable Rotation Estimation with Transformers
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2512.18784