Learning Multi-Scene Absolute Pose Regression with Transformers

Abstract

Absolute camera pose regressors estimate the position and orientation of a camera from the captured image alone. Typically, a convolutional backbone with a multi-layer perceptron head is trained with images and pose labels to embed a single reference scene at a time. Recently, this scheme was extended for learning multiple scenes by replacing the MLP head with a set of fully connected layers. In this work, we propose to learn multi-scene absolute camera pose regression with Transformers, where encoders are used to aggregate activation maps with self-attention and decoders transform latent features and scenes encoding into candidate pose predictions. This mechanism allows our model to focus on general features that are informative for localization while embedding multiple scenes in parallel. We evaluate our method on commonly benchmarked indoor and outdoor datasets and show that it surpasses both multi-scene and state-of-the-art single-scene absolute pose regressors. We make our code publicly available from here.

ViewFormer: NeRF-Free Neural Rendering from Few Images Using Transformers

“…Our approach performs similar to existing APR and RPR techniques that also use only a single forward pass in a network [1,8,30,60], but worse than iterative approaches such as [19] or methods that use more densely spaced synthetic views as additional input [41]. Note that these approaches that do not use 3D scene geometry are less accurate than state-of-the-art methods based on 2D-3D correspondences [7,56,58].…”

“…Since this task can be considered an "inverse" of the novel view synthesis task [70], we consider the ability to perform both tasks via the same model to be an intriguing property. Even though the localization results are not yet competitive with state-of-the-art localization pipelines, we achieve a similar level of pose accuracy as comparable methods such as [1,60].…”

“…Pose regression methods train a convolutional neural network (CNN) to regress the camera pose of an input image. There are two categories: absolute pose regression (APR) methods [5,8,14,28,30,33,41,60] and relative pose regression (RPR) methods [1,19,31,33,39]. It was shown [59] that APR is often not (much) more accurate than IR.…”