Depth Pro: Sharp Monocular Metric Depth in Less Than a Second

We present a foundation model for zero-shot metric monocular depth estimation. Our model, Depth Pro, synthesizes high-resolution depth maps with unparalleled sharpness and high-frequency details. The predictions are metric, with absolute scale, without relying on the availability of metadata such as camera intrinsics. And the model is fast, producing a 2.25-megapixel depth map in 0.3 seconds on a standard GPU. These characteristics are enabled by a number of technical contributions, including an efficient multi-scale vision transformer for dense prediction, a training protocol that combines real and synthetic datasets to achieve high metric accuracy alongside fine boundary tracing, dedicated evaluation metrics for boundary accuracy in estimated depth maps, and state-of-the-art focal length estimation from a single image. Extensive experiments analyze specific design choices and demonstrate that Depth Pro outperforms prior work along multiple dimensions. We release code and weights at https://github.com/apple/ml-depth-pro

Are Semi-Dense Detector-Free Methods Good at Matching Local Features?

Semi-dense detector-free approaches (SDF), such as LoFTR, are currently among the most popular image matching methods. While SDF methods are trained to establish correspondences between two images, their performances are almost exclusively evaluated using relative pose estimation metrics. Thus, the link between their ability to establish correspondences and the quality of the resulting estimated pose has thus far received little attention. This paper is a first attempt to study this link. We start with proposing a novel structured attention-based image matching architecture (SAM). It allows us to show a counter-intuitive result on two datasets (MegaDepth and HPatches): on the one hand SAM either outperforms or is on par with SDF methods in terms of pose/homography estimation metrics, but on the other hand SDF approaches are significantly better than SAM in terms of matching accuracy. We then propose to limit the computation of the matching accuracy to textured regions, and show that in this case SAM often surpasses SDF methods. Our findings highlight a strong correlation between the ability to establish accurate correspondences in textured regions and the accuracy of the resulting estimated pose/homography. Our code will be made available.

Nerfels: Renderable Neural Codes for Improved Camera Pose Estimation

This paper presents a framework that combines traditional keypoint-based camera pose optimization with an invertible neural rendering mechanism. Our proposed 3D scene representation, Nerfels, is locally dense yet globally sparse. As opposed to existing invertible neural rendering systems which overfit a model to the entire scene, we adopt a feature-driven approach for representing scene-agnostic, local 3D patches with renderable codes. By modelling a scene only where local features are detected, our framework effectively generalizes to unseen local regions in the scene via an optimizable code conditioning mechanism in the neural renderer, all while maintaining the low memory footprint of a sparse 3D map representation. Our model can be incorporated to existing state-of-the-art hand-crafted and learned local feature pose estimators, yielding improved performance when evaluating on ScanNet for wide camera baseline scenarios.

Visual Correspondence Hallucination: Towards Geometric Reasoning

Given a pair of partially overlapping source and target images and a keypoint in the source image, the keypoint's correspondent in the target image can be either visible, occluded or outside the field of view. Local feature matching methods are only able to identify the correspondent's location when it is visible, while humans can also hallucinate its location when it is occluded or outside the field of view through geometric reasoning. In this paper, we bridge this gap by training a network to output a peaked probability distribution over the correspondent's location, regardless of this correspondent being visible, occluded, or outside the field of view. We experimentally demonstrate that this network is indeed able to hallucinate correspondences on unseen pairs of images. We also apply this network to a camera pose estimation problem and find it is significantly more robust than state-of-the-art local feature matching-based competitors.

Back to the Feature: Learning Robust Camera Localization from Pixels to Pose

Camera pose estimation in known scenes is a 3D geometry task recently tackled by multiple learning algorithms. Many regress precise geometric quantities, like poses or 3D points, from an input image. This either fails to generalize to new viewpoints or ties the model parameters to a specific scene. In this paper, we go Back to the Feature: we argue that deep networks should focus on learning robust and invariant visual features, while the geometric estimation should be left to principled algorithms. We introduce PixLoc, a scene-agnostic neural network that estimates an accurate 6-DoF pose from an image and a 3D model. Our approach is based on the direct alignment of multiscale deep features, casting camera localization as metric learning. PixLoc learns strong data priors by end-to-end training from pixels to pose and exhibits exceptional generalization to new scenes by separating model parameters and scene geometry. The system can localize in large environments given coarse pose priors but also improve the accuracy of sparse feature matching by jointly refining keypoints and poses with little overhead. The code will be publicly available at https://github.com/cvg/pixloc.

Neural Reprojection Error: Merging Feature Learning and Camera Pose Estimation

Absolute camera pose estimation is usually addressed by sequentially solving two distinct subproblems: First a feature matching problem that seeks to establish putative 2D-3D correspondences, and then a Perspective-n-Point problem that minimizes, with respect to the camera pose, the sum of so-called Reprojection Errors (RE). We argue that generating putative 2D-3D correspondences 1) leads to an important loss of information that needs to be compensated as far as possible, within RE, through the choice of a robust loss and the tuning of its hyperparameters and 2) may lead to an RE that conveys erroneous data to the pose estimator. In this paper, we introduce the Neural Reprojection Error (NRE) as a substitute for RE. NRE allows to rethink the camera pose estimation problem by merging it with the feature learning problem, hence leveraging richer information than 2D-3D correspondences and eliminating the need for choosing a robust loss and its hyperparameters. Thus NRE can be used as training loss to learn image descriptors tailored for pose estimation. We also propose a coarse-to-fine optimization method able to very efficiently minimize a sum of NRE terms with respect to the camera pose. We experimentally demonstrate that NRE is a good substitute for RE as it significantly improves both the robustness and the accuracy of the camera pose estimate while being computationally and memory highly efficient. From a broader point of view, we believe this new way of merging deep learning and 3D geometry may be useful in other computer vision applications.

S2DNet: Learning Accurate Correspondences for Sparse-to-Dense Feature Matching

Establishing robust and accurate correspondences is a fundamental backbone to many computer vision algorithms. While recent learning-based feature matching methods have shown promising results in providing robust correspondences under challenging conditions, they are often limited in terms of precision. In this paper, we introduce S2DNet, a novel feature matching pipeline, designed and trained to efficiently establish both robust and accurate correspondences. By leveraging a sparse-to-dense matching paradigm, we cast the correspondence learning problem as a supervised classification task to learn to output highly peaked correspondence maps. We show that S2DNet achieves state-of-the-art results on the HPatches benchmark, as well as on several long-term visual localization datasets.

Sparse-to-Dense Hypercolumn Matching for Long-Term Visual Localization

We propose a novel approach to feature point matching, suitable for robust and accurate outdoor visual localization in long-term scenarios. Given a query image, we first match it against a database of registered reference images, using recent retrieval techniques. This gives us a first estimate of the camera pose. To refine this estimate, like previous approaches, we match 2D points across the query image and the retrieved reference image. This step, however, is prone to fail as it is still very difficult to detect and match sparse feature points across images captured in potentially very different conditions. Our key contribution is to show that we need to extract sparse feature points only in the retrieved reference image: We then search for the corresponding 2D locations in the query image exhaustively. This search can be performed efficiently using convolutional operations, and robustly by using hypercolumn descriptors, i.e. image features computed for retrieval. We refer to this method as Sparse-to-Dense Hypercolumn Matching. Because we know the 3D locations of the sparse feature points in the reference images thanks to an offline reconstruction stage, it is then possible to accurately estimate the camera pose from these matches. Our experiments show that this method allows us to outperform the state-of-the-art on several challenging outdoor datasets.

Efficient Condition-based Representations for Long-Term Visual Localization

We propose an approach to localization from images that is designed to explicitly handle the strong variations in appearance happening when capturing conditions change throughout the day or across seasons. As revealed by recent long-term localization benchmarks, both traditional feature-based and retrieval-based approaches still struggle to handle such changes. Our novel retrieval-based method introduces condition-specific sub-networks allowing the computation of global image descriptors that are explicitly dependent of the capturing conditions. We compare our approach to previous localization methods on very recent challenging benchmarks, and observe that our method outperforms them by a large margin in case of day-night variation, where repeatable feature points cannot be identified or matched.