Paying Attention to Activation Maps in Camera Pose Regression

TL;DR

This paper introduces an attention-based transformer approach for camera pose regression that leverages activation maps to improve accuracy, achieving state-of-the-art results including sub-meter precision in outdoor scenes.

Contribution

It proposes a novel transformer-based method using activation maps for pose regression, enabling focused attention on spatial features for position and orientation estimation.

Findings

Outperforms existing pose regressors on multiple benchmarks.

Achieves sub-meter accuracy in outdoor scenes.

Demonstrates the effectiveness of attention mechanisms in pose estimation.

Abstract

Camera pose regression methods apply a single forward pass to the query image to estimate the camera pose. As such, they offer a fast and light-weight alternative to traditional localization schemes based on image retrieval. Pose regression approaches simultaneously learn two regression tasks, aiming to jointly estimate the camera position and orientation using a single embedding vector computed by a convolutional backbone. We propose an attention-based approach for pose regression, where the convolutional activation maps are used as sequential inputs. Transformers are applied to encode the sequential activation maps as latent vectors, used for camera pose regression. This allows us to pay attention to spatially-varying deep features. Using two Transformer heads, we separately focus on the features for camera position and orientation, based on how informative they are per task. Our proposed approach is shown to compare favorably to contemporary pose regressors schemes and achieves state-of-the-art accuracy across multiple outdoor and indoor benchmarks. In particular, to the best of our knowledge, our approach is the only method to attain sub-meter average accuracy across outdoor scenes. We make our code publicly available from here.