Visual triage: A bag-of-words experience selector for long-term visual route following

MacTavish, Kirk; Paton, Michael; Barfoot, Timothy D.

doi:10.1109/icra.2017.7989238

Cited by 21 publications

(17 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Please note that a direct comparison of our appearance-based landmark selection performance with the most related works (Linegar et al, 2015;Mactavish et al, 2017) is inherently difficult, as the underlying mapping framework and visual feature representations are fundamentally different, and the selection of relevant data on the level of individual landmarks constitutes a unique feature of our method. With the ranking function f MRS , as introduced in Section 5.3, we aim at comparing our method with the performance that is to be expected with an "experience-based" mapping framework, which BÜRKI ET AL.…”

Section: Discussionmentioning

confidence: 99%

Appearance‐based landmark selection for visual localization

Bürki

Cadena

Gilitschenski³

et al. 2019

Journal of Field Robotics

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Appearance‐based landmark selection for visual localization

Bürki

Cadena

Gilitschenski³

et al. 2019

Journal of Field Robotics

View full text Add to dashboard Cite

show abstract

“…In previous work (MacTavish, Paton & Barfoot, ), we introduced an experience‐triaging technique based on a sliding‐window bag‐of‐words (BoW). This study leveraged success from place recognition (MacTavish & Barfoot, ; MacTavish, Paton, & Barfoot, ), with the intention of quickly searching for similar experiences rather than places.…”

Section: Related Workmentioning

confidence: 99%

“…An example of a content‐based recommender applied to our problem is introduced by MacTavish et al (). Landmarks are quantized into words in a visual vocabulary based on their visual descriptor.…”

Section: Related Workmentioning

confidence: 99%

Selective memory: Recalling relevant experience for long‐term visual localization

MacTavish

Paton

Barfoot

2018

Journal of Field Robotics

Self Cite

View full text Add to dashboard Cite

Visual navigation is a key enabling technology for autonomous mobile vehicles. The ability to provide large-scale, long-term navigation using low-cost, low-power vision sensors is appealing for industrial applications. A crucial requirement for long-term navigation systems is the ability to localize in environments whose appearance is constantly changing over time-due to lighting, weather, seasons, and physical changes. This paper presents a multiexperience localization (MEL) system that uses a powerful map representation-storing every visual experience in layers-that does not make assumptions about underlying appearance modalities and generators. Our localization system provides real-time performance by selecting online, a subset of experiences against which to localize. We achieve this task through a novel experience-triage algorithm based on collaborative filtering, which selects experiences relevant to the live view, outperforming competing techniques. Based on classical memory-based recommender systems, this technique also enables landmarklevel recommendations, is entirely online, and requires no training data. We demonstrate the capabilities of the MEL system in the context of long-term autonomous path following in unstructured outdoor environments with a challenging 100-day field experiment through day, night, snow, spring, and summer. We furthermore provide offline analysis comparing our system to several state-of-theart alternatives. We show that the combination of the novel methods presented in this paper enable full use of incredibly rich multiexperience maps, opening the door to robust long-term visual localization.

show abstract

“…However, this work is sensitive to changes of viewpoint and hence requires the initial position to be similar. To handle graduallyaccumulating environmental changes, [12] employs a 'multiexperience' method for VTR, which was introduced in [13]. Krajkik et al [14] examines combinations of feature detectors and descriptors for VTR, favouring the STAR feature detector and GRIEF feature descriptor for good accuracy with computational speed.…”

Section: Related Workmentioning

confidence: 99%

Robust UAV Visual Teach and Repeat Using Only Sparse Semantic Object Features

Toudeshki

Shamshirdar

2018

2018 15th Conference on Computer and Robot Vision (CRV)

View full text Add to dashboard Cite

We demonstrate the use of semantic object detections as robust features for Visual Teach and Repeat (VTR). Recent CNN-based object detectors are able to reliably detect objects of tens or hundreds of categories in video at frame rates. We show that such detections are repeatable enough to use as landmarks for VTR, without any low-level image features. Since object detections are highly invariant to lighting and surface appearance changes, our VTR can cope with global lighting changes and local movements of the landmark objects. In the teaching phase we build a series of compact scene descriptors: a list of detected object labels and their image-plane locations. In the repeating phase, we use Seq-SLAM-like relocalization to identify the most similar learned scene, then use a motion control algorithm based on the funnel lane theory to navigate the robot along the previously piloted trajectory.We evaluate the method on a commodity UAV, examining the robustness of the algorithm to new viewpoints, lighting conditions, and movements of landmark objects. The results suggest that semantic object features could be useful due to their invariance to superficial appearance changes compared to low-level image features.

show abstract

Visual triage: A bag-of-words experience selector for long-term visual route following

Cited by 21 publications

References 15 publications

Appearance‐based landmark selection for visual localization

Appearance‐based landmark selection for visual localization

Selective memory: Recalling relevant experience for long‐term visual localization

Robust UAV Visual Teach and Repeat Using Only Sparse Semantic Object Features

Contact Info

Product

Resources

About