Appearance-based Active, Monocular, Dense Reconstruction for Micro Aerial Vehicles

Förster, Christian; Pizzoli, Matia; Scaramuzza, Davide

doi:10.15607/rss.2014.x.029

Cited by 45 publications

(41 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Research on the Next-Best-View problem and conceptually similar problems in Active Vision dates back several decades Aloimonos et al (1988), Bajcsy (1988) but remains an active area of research Forster et al (2014). The most frequently referenced surveys of the field include an overview of early approaches by Scott et al (2003) and an overview of more recent work by Chen et al (2011).…”

Section: Related Workmentioning

confidence: 99%

“…They rely on knowledge of the geometry and appearance of the object, which may not be available in many real world scenarios. Non-model based approaches use relaxed assumptions about the structure of the object, but the required information for planning the next best view must be estimated online based on the gathered data Banta et al (1995), Forster et al (2014). We utilize a non-model based approach since we do not assume anything about the object aside from its spatial bounds.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A comparison of volumetric information gain metrics for active 3D object reconstruction

et al. 2017

Self Cite

View full text Add to dashboard Cite

In this paper, we investigate the following question: when performing next best view selection for volumetric 3D reconstruction of an object by a mobile robot equipped with a dense (camera-based) depth sensor, what formulation of information gain is best? To address this question, we propose several new ways to quantify the volumetric information (VI) contained in the voxels of a probabilistic volumetric map, and compare them to the state of the art with extensive simulated experiments. Our proposed formulations incorporate factors such as visibility likelihood and the likelihood of seeing new parts of the object. The results of our experiments allow us to draw some clear conclusions about the VI formulations that are most effective in different mobile-robot reconstruction scenarios. To the best of our knowledge, this is the first comparative survey of VI formulation performance for active 3D object reconstruction. Additionally, our modular software framework is adaptable to other robotic platforms and general reconstruction problems, and we release it open source for autonomous reconstruction tasks. Abstract In this paper, we investigate the following question: when performing next best view selection for volumetric 3D reconstruction of an object by a mobile robot equipped with a dense (camera-based) depth sensor, what formulation of information gain is best? To address this question, we propose several new ways to quantify the volumetric information (VI) contained in the voxels of a probabilistic volumetric map, and compare them to the state of the art with extensive simulated experiments. Our proposed formulations incorporate factors such as visibility likelihood and the likelihood of seeing new parts of the object. The results of our experiments allow us to draw some clear conclusions about the VI formulations that are most effective in different mobile-robot reconstruction scenarios. To the best of our knowledge, this is the first comparative survey of VI formulation performance for active 3D object reconstruction. Additionally, our modular software framework is adaptable to other robotic platforms and general reconstruction problems, and we release it open source for autonomous reconstruction tasks.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

A comparison of volumetric information gain metrics for active 3D object reconstruction

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, Forster et al [5] use depth uncertainty to estimate the best areas of the map to explore but are limited to relatively simple scenes. Hoppe et al [10] create a full network of poses for an Unmanned Aerial Vehicle (UAV), but assumes prior knowledge of the environment.…”

Section: Next-best View (Nbv) Estimationmentioning

confidence: 99%

“…On the other hand, very large values start to suffer from increasing depth error due to the relatively narrow baseline. Choosing values of α ∈ [5,7], corresponding to a vergence angle of around 9 • degrees, achieves high coverage while minimising the average error. It is important to note that these results are not dependent on the absolute values of depth, baseline or vergence.…”

Section: Parameter Explorationmentioning

confidence: 99%

Next-Best Stereo: Extending Next-Best View Optimisation For Collaborative Sensors

Maldonado¹,

Hadfield²,

Pugeault³

et al. 2016

Procedings of the British Machine Vision Conference 2016

View full text Add to dashboard Cite

Most 3D reconstruction approaches passively optimise over all data, exhaustively matching pairs, rather than actively selecting data to process. This is costly both in terms of time and computer resources, and quickly becomes intractable for large datasets.This work proposes an approach to intelligently filter large amounts of data for 3D reconstructions of unknown scenes using monocular cameras. Our contributions are twofold: First, we present a novel approach to efficiently optimise the Next-Best View (NBV) in terms of accuracy and coverage using partial scene geometry. Second, we extend this to intelligently selecting stereo pairs by jointly optimising the baseline and vergence to find the NBV's best stereo pair to perform reconstruction. Both contributions are extremely efficient, taking 0.8ms and 0.3ms per pose, respectively.Experimental evaluation shows that the proposed method allows efficient selection of stereo pairs for reconstruction, such that a dense model can be obtained with only a small number of images. Once a complete model has been obtained, the remaining computational budget is used to intelligently refine areas of uncertainty, achieving results comparable to state-of-the-art batch approaches on the Middlebury dataset, using as little as 3.8% of the views.

show abstract

“…Recently it has been applied successfully to dense 3D reconstruction [8], image-based rendering [7], image matching [32] and even used with RGBD data for object detection [12].…”

Section: Introductionmentioning

confidence: 99%

Matchability Prediction for Full-Search Template Matching Algorithms

Peñate-Sánchez

Porzi

Moreno-Noguer

2015

2015 International Conference on 3D Vision

View full text Add to dashboard Cite

While recent approaches have shown that it is possible to do template matching by exhaustively scanning the parameter space, the resulting algorithms are still quite demanding. In this paper we alleviate the computational load of these algorithms by proposing an efficient approach for predicting the matchability of a template, before it is actually performed. This avoids large amounts of unnecessary computations. We learn the matchability of templates by using dense convolutional neural network descriptors that do not require ad-hoc criteria to characterize a template. By using deep learning descriptions of patches we are able to predict matchability over the whole image quite reliably. We will also show how no specific training data is required to solve problems like panorama stitching in which you usually require data from the scene in question. Due to the highly parallelizable nature of this tasks we offer an efficient technique with a negligible computational cost at test time.

show abstract

Appearance-based Active, Monocular, Dense Reconstruction for Micro Aerial Vehicles

Cited by 45 publications

References 28 publications

A comparison of volumetric information gain metrics for active 3D object reconstruction

A comparison of volumetric information gain metrics for active 3D object reconstruction

Next-Best Stereo: Extending Next-Best View Optimisation For Collaborative Sensors

Matchability Prediction for Full-Search Template Matching Algorithms

Contact Info

Product

Resources

About