Open-RadVLAD: fast shift and rotation invariant radar place recognition
—Radar place recognition often involves encoding a live scan as a vector and matching this vector to a database in order to recognise that the vehicle is in a location that it has visited before. Radar is inherently robust to lighting or weather conditions, but place recognition with this sensor is...
| Main Authors: | , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
IEEE
2024
|
| _version_ | 1797113272766300160 |
|---|---|
| author | Gadd, M Newman, P |
| author_facet | Gadd, M Newman, P |
| author_sort | Gadd, M |
| collection | OXFORD |
| description | —Radar place recognition often involves encoding a
live scan as a vector and matching this vector to a database in
order to recognise that the vehicle is in a location that it has
visited before. Radar is inherently robust to lighting or weather
conditions, but place recognition with this sensor is still affected
by: (1) viewpoint variation, i.e. translation and rotation, (2)
sensor artefacts or “noises”. For 360◦
scanning radar, rotation is
readily dealt with by in some way aggregating across azimuths.
Also, we argue in this work that it is more critical to deal
with the richness of representation and sensor noises than it
is to deal with translational invariance – particularly in urban
driving where vehicles predominantly follow the same lane when
repeating a route. In our method, for computational efficiency,
we use only the polar representation. For partial translation
invariance and robustness to signal noise, we use only a onedimensional Fourier Transform along radial returns. We also
achieve rotational invariance and a very discriminative descriptor
space by building a vector of locally aggregated descriptors. Our
method is more comprehensively tested than all prior radar
place recognition work – over an exhaustive combination of
all 870 pairs of trajectories from 30 Oxford Radar RobotCar
Dataset sequences (each ≈10 km). Code and detailed results
are provided at github.com/mttgdd/open-radvlad, as an
open implementation and benchmark for future work in this area.
We achieve a median of 91.52 % in Recall@1, outstripping the
69.55 % for the only other open implementation, RaPlace, and
at a fraction of its computational cost (relying on fewer integral
transforms e.g. Radon, Fourier, and inverse Fourier). |
| first_indexed | 2024-03-07T08:29:36Z |
| format | Conference item |
| id | oxford-uuid:b11b1f95-4c04-4175-9a46-6ee05e84377b |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-04-23T08:26:11Z |
| publishDate | 2024 |
| publisher | IEEE |
| record_format | dspace |
| spelling | oxford-uuid:b11b1f95-4c04-4175-9a46-6ee05e84377b2024-04-15T09:31:39ZOpen-RadVLAD: fast shift and rotation invariant radar place recognitionConference itemhttp://purl.org/coar/resource_type/c_5794uuid:b11b1f95-4c04-4175-9a46-6ee05e84377bEnglishSymplectic ElementsIEEE2024Gadd, MNewman, P—Radar place recognition often involves encoding a live scan as a vector and matching this vector to a database in order to recognise that the vehicle is in a location that it has visited before. Radar is inherently robust to lighting or weather conditions, but place recognition with this sensor is still affected by: (1) viewpoint variation, i.e. translation and rotation, (2) sensor artefacts or “noises”. For 360◦ scanning radar, rotation is readily dealt with by in some way aggregating across azimuths. Also, we argue in this work that it is more critical to deal with the richness of representation and sensor noises than it is to deal with translational invariance – particularly in urban driving where vehicles predominantly follow the same lane when repeating a route. In our method, for computational efficiency, we use only the polar representation. For partial translation invariance and robustness to signal noise, we use only a onedimensional Fourier Transform along radial returns. We also achieve rotational invariance and a very discriminative descriptor space by building a vector of locally aggregated descriptors. Our method is more comprehensively tested than all prior radar place recognition work – over an exhaustive combination of all 870 pairs of trajectories from 30 Oxford Radar RobotCar Dataset sequences (each ≈10 km). Code and detailed results are provided at github.com/mttgdd/open-radvlad, as an open implementation and benchmark for future work in this area. We achieve a median of 91.52 % in Recall@1, outstripping the 69.55 % for the only other open implementation, RaPlace, and at a fraction of its computational cost (relying on fewer integral transforms e.g. Radon, Fourier, and inverse Fourier). |
| spellingShingle | Gadd, M Newman, P Open-RadVLAD: fast shift and rotation invariant radar place recognition |
| title | Open-RadVLAD: fast shift and rotation invariant radar place recognition |
| title_full | Open-RadVLAD: fast shift and rotation invariant radar place recognition |
| title_fullStr | Open-RadVLAD: fast shift and rotation invariant radar place recognition |
| title_full_unstemmed | Open-RadVLAD: fast shift and rotation invariant radar place recognition |
| title_short | Open-RadVLAD: fast shift and rotation invariant radar place recognition |
| title_sort | open radvlad fast shift and rotation invariant radar place recognition |
| work_keys_str_mv | AT gaddm openradvladfastshiftandrotationinvariantradarplacerecognition AT newmanp openradvladfastshiftandrotationinvariantradarplacerecognition |