Relative Pose from Deep Learned Depth and a Single Affine Correspondence

Eichhardt, Ivan; Baráth, Dániel

doi:10.1007/978-3-030-58610-2_37

Cited by 12 publications

(8 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The relative camera pose estimation between adjacent cameras is an essential low-level task for many computer vision applications, such as simultaneous localization and mapping (SLAM) [26][27][28], depth estimation [29][30][31][32], and visual odometry [33][34][35]. Several attempts have been made on the topic [36][37][38][39][40] due to its widespread usage. Nonetheless, much progress has been driven by studies on minimal solvers [36][37][38] or the practice of affine correspondence [39,40].…”

Section: Camera Pose Estimationmentioning

confidence: 99%

“…Several attempts have been made on the topic [36][37][38][39][40] due to its widespread usage. Nonetheless, much progress has been driven by studies on minimal solvers [36][37][38] or the practice of affine correspondence [39,40]. Although the relative camera pose estimation and our framework extract accurate camera poses, our framework seeks to exploit a priori information on camera arrangement and extract camera poses located in the world coordinate system of the complete image set.…”

Section: Camera Pose Estimationmentioning

confidence: 99%

See 1 more Smart Citation

Camera pose estimation framework for array‐structured images

et al. 2022

View full text Add to dashboard Cite

Despite the significant progress in camera pose estimation and structure‐from‐motion reconstruction from unstructured images, methods that exploit a priori information on camera arrangements have been overlooked. Conventional state‐of‐the‐art methods do not exploit the geometric structure to recover accurate camera poses from a set of patch images in an array for mosaic‐based imaging that creates a wide field‐of‐view image by sewing together a collection of regular images. We propose a camera pose estimation framework that exploits the array‐structured image settings in each incremental reconstruction step. It consists of the two‐way registration, the 3D point outlier elimination and the bundle adjustment with a constraint term for consistent rotation vectors to reduce reprojection errors during optimization. We demonstrate that by using individual images' connected structures at different camera pose estimation steps, we can estimate camera poses more accurately from all structured mosaic‐based image sets, including omnidirectional scenes.

show abstract

Section: Camera Pose Estimationmentioning

confidence: 99%

Section: Camera Pose Estimationmentioning

confidence: 99%

Camera pose estimation framework for array‐structured images

et al. 2022

View full text Add to dashboard Cite

show abstract

“…When the relative pose has a motion prior, there are simple and efficient solvers. For example, there are many customized solvers when the rotation axis is known [21], the motion is under planar motion [21], [22], or the depth of feature points is known [53].…”

Section: Related Workmentioning

confidence: 99%

On Relative Pose Recovery for Multi-Camera Systems

Zhao¹,

Guan²

2021

Preprint

View full text Add to dashboard Cite

The point correspondence (PC) and affine correspondence (AC) are widely used for relative pose estimation. An AC consists of a PC across two views and an affine transformation between the small patches around this PC. Previous work demonstrates that one AC generally provides three independent constraints for relative pose estimation. For multi-camera systems, there is still not any AC-based minimal solver for general relative pose estimation. To deal with this problem, we propose a complete solution to relative pose estimation from two ACs for multi-camera systems, consisting of a series of minimal solvers. The solver generation in our solution is based on Cayley or quaternion parameterization for rotation and hidden variable technique to eliminate translation. This solver generation method is also naturally applied to relative pose estimation from PCs, resulting in a new six-point method for multi-camera systems. A few extensions are made, including relative pose estimation with known rotation angle and/or with unknown focal lengths. Extensive experiments demonstrate that the proposed AC-based solvers and PC-based solvers are effective and efficient on synthetic and real-world datasets.

show abstract

“…Besides the special case considerations, additional constraints can also come from running other algorithms, like monocular depth estimation. Such a constraint could reduce the required number of matches from two affine correspondences to a single one for the calibrated camera case [103].…”

Section: Wide Baseline Stereomentioning

confidence: 99%

Learning and Crafting for the Wide Multiple Baseline Stereo

Mishkin¹

2021

Preprint

View full text Add to dashboard Cite

This thesis introduces the wide multiple baseline stereo (WxBS) problem. WxBS, a generalization of the standard wide baseline stereo problem, considers the matching of images that simultaneously differ in more than one image acquisition factor such as viewpoint, illumination, sensor type, or where object appearance changes significantly, e.g., over time. A new dataset with the ground truth, evaluation metric and baselines has been introduced.The thesis presents the following improvements of the WxBS pipeline. (i) A loss function, called HardNeg, for learning a local image descriptor that relies on hard negative mining within a mini-batch and on the maximization of the distance between the closest positive and the closest negative patches. (ii) The descriptor trained with the HardNeg loss, called HardNet, is compact and shows state-of-the-art performance in standard matching, patch verification and retrieval benchmarks. (iii) A method for learning the affine shape, orientation, and potentially other parameters related to geometric and appearance properties of local features. (iv) A tentative correspondences generation strategy which generalizes the standard first to second closest distance ratio is presented. The selection strategy, which shows performance superior to the standard method, is applicable to either hard-engineered descriptors like SIFT, LIOP, and MROGH or deeply learned like HardNet. (v) A feedback loop is introduced for the two-view matching problem, resulting in MODS -matching with on-demand view synthesis -algorithm. MODS is an algorithm that handles a viewing angle difference even larger than the previous state-of-the-art ASIFT algorithm, without a significant increase of computational cost over "standard" wide and narrow baseline approaches.Last, but not least, a comprehensive benchmark for local features and robust estimation algorithms is introduced. The modular structure of its pipeline allows easy integration, configuration, and combination of methods and heuristics. AbstraktTato práce představuje problém wide multiple baseline stereo (WxBS), který je zobecněním standardního problému wide baseline stereo. WxBS se zabývá párováním obrázků, které se současně liší více než v jednom faktoru získání obrázku, jako je stanoviště a zorný úhel, osvětlení, typ senzoru, nebo kde se vzhled objektu zásadně mění, např. v průběhu času. Byl představen nový dataset s anotacemi, evaluační metrikou a výchozími metodami.Disertační práce představuje následující vylepšení algoritmu WxBS: Ztrátovou funkci (i) pro učení lokálního deskriptoru obrázku založeného na výběru obtížných negativních příkladů v minibatchi a na maximalizaci rozdílu vzdálenosti od nejbližšího pozitivního a nejbližšího negativního příkladu. Výsledný deskriptor (ii), nazvaný HardNet, který je kompaktní a dosahuje vynikajících výsledků v standardním párování, ověřování malých částí obrázků a vyhledávání obrázků. Metodu pro učení afinního tvaru, orientace a potenciálně dalších parametrů souvisejících s geometrií a vzhledem lokálních příznaků (iii...

show abstract

Relative Pose from Deep Learned Depth and a Single Affine Correspondence

Cited by 12 publications

References 55 publications

Camera pose estimation framework for array‐structured images

Camera pose estimation framework for array‐structured images

On Relative Pose Recovery for Multi-Camera Systems

Learning and Crafting for the Wide Multiple Baseline Stereo

Contact Info

Product

Resources

About